diff --git a/.gitignore b/.gitignore
index 4524963..41846df 100644
--- a/.gitignore
+++ b/.gitignore
@@ -21,6 +21,8 @@ __pycache__/
 *.py[cod]
 *$py.class
 
+# pycharm
+**/*.idea/*
 # C extensions
 *.so
 
diff --git a/3_cascade/1_sanity_check_run.ipynb b/3_cascade/1_sanity_check_run.ipynb
index 36a9ab2..bb9a77a 100644
--- a/3_cascade/1_sanity_check_run.ipynb
+++ b/3_cascade/1_sanity_check_run.ipynb
@@ -23,42 +23,12 @@
     "import matplotlib.pyplot as plt\n",
     "from ase.db import connect\n",
     "\n",
-    "from cascade.agents.db_orm import TrajectoryDB"
+    "from cascade.agents.db_orm import TrajectoryDB, DBTrajectoryFrame"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 2,
-   "id": "040bfdb3-04df-4857-9692-4474c3109d09",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "\u001b[0;31mInit signature:\u001b[0m \u001b[0mTrajectoryDB\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdb_url\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'str'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogger\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'Optional[logging.Logger]'\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-       "\u001b[0;31mDocstring:\u001b[0m      Wrapper for the database representations of trajectories and chunks\n",
-       "\u001b[0;31mInit docstring:\u001b[0m\n",
-       "Initialize the trajectory database manager\n",
-       "\n",
-       "Args:\n",
-       "    db_url: PostgreSQL connection URL (e.g., 'postgresql://user:pass@host:port/dbname')\n",
-       "    logger: Optional logger for tracking engine creation\n",
-       "\u001b[0;31mFile:\u001b[0m           ~/repos/cascade/cascade/agents/db_orm.py\n",
-       "\u001b[0;31mType:\u001b[0m           type\n",
-       "\u001b[0;31mSubclasses:\u001b[0m     "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "?TrajectoryDB"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
    "id": "d83efa71-637a-4607-9310-bdab4d9a1970",
    "metadata": {},
    "outputs": [],
@@ -70,7 +40,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "id": "e155bfeb-37f7-466d-9240-7f34f4b018b9",
    "metadata": {},
    "outputs": [],
@@ -80,7 +50,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "id": "46b842fc-2638-4802-b3bb-d4070a95cc1c",
    "metadata": {},
    "outputs": [
@@ -115,43 +85,43 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2026-02-05 14:21:25.016871-06:00</td>\n",
-       "      <td>2026-02-05 14:24:37.524925-06:00</td>\n",
-       "      <td>4</td>\n",
-       "      <td>4</td>\n",
+       "      <td>2026.03.24-18:25:55-03607e</td>\n",
+       "      <td>2026-03-24 13:25:55.484936-05:00</td>\n",
+       "      <td>2026-03-24 13:34:43.160574-05:00</td>\n",
+       "      <td>16</td>\n",
+       "      <td>16</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>2026.02.05-20:19:17-144915</td>\n",
-       "      <td>2026-02-05 14:19:18.219930-06:00</td>\n",
-       "      <td>2026-02-05 14:20:07.585033-06:00</td>\n",
-       "      <td>2</td>\n",
-       "      <td>2</td>\n",
+       "      <td>2026.03.24-18:20:18-8992ba</td>\n",
+       "      <td>2026-03-24 13:20:18.221102-05:00</td>\n",
+       "      <td>2026-03-24 13:24:23.428353-05:00</td>\n",
+       "      <td>16</td>\n",
+       "      <td>16</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>2026.02.05-20:17:15-144915</td>\n",
-       "      <td>2026-02-05 14:17:15.790865-06:00</td>\n",
-       "      <td>2026-02-05 14:17:55.001234-06:00</td>\n",
-       "      <td>2</td>\n",
-       "      <td>2</td>\n",
+       "      <td>2026.03.24-18:11:53-8992ba</td>\n",
+       "      <td>2026-03-24 13:11:53.374532-05:00</td>\n",
+       "      <td>2026-03-24 13:12:42.391336-05:00</td>\n",
+       "      <td>16</td>\n",
+       "      <td>4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>2026.02.05-20:12:41-c0e31c</td>\n",
-       "      <td>2026-02-05 14:12:41.917551-06:00</td>\n",
-       "      <td>2026-02-05 14:13:44.568926-06:00</td>\n",
-       "      <td>4</td>\n",
-       "      <td>1</td>\n",
+       "      <td>2026.03.24-17:51:33-8992ba</td>\n",
+       "      <td>2026-03-24 12:51:33.144422-05:00</td>\n",
+       "      <td>2026-03-24 12:52:17.679446-05:00</td>\n",
+       "      <td>16</td>\n",
+       "      <td>3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>2026.02.05-20:00:22-c0e31c</td>\n",
-       "      <td>2026-02-05 14:00:22.867890-06:00</td>\n",
-       "      <td>2026-02-05 14:01:13.190946-06:00</td>\n",
+       "      <td>2026.03.24-17:47:25-c80e5e</td>\n",
+       "      <td>2026-03-24 12:47:25.169991-05:00</td>\n",
+       "      <td>2026-03-24 12:49:43.698338-05:00</td>\n",
+       "      <td>4</td>\n",
        "      <td>4</td>\n",
-       "      <td>0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -159,21 +129,21 @@
       ],
       "text/plain": [
        "                       run_id                     first_created  \\\n",
-       "0  2026.02.05-20:21:24-c0e31c  2026-02-05 14:21:25.016871-06:00   \n",
-       "1  2026.02.05-20:19:17-144915  2026-02-05 14:19:18.219930-06:00   \n",
-       "2  2026.02.05-20:17:15-144915  2026-02-05 14:17:15.790865-06:00   \n",
-       "3  2026.02.05-20:12:41-c0e31c  2026-02-05 14:12:41.917551-06:00   \n",
-       "4  2026.02.05-20:00:22-c0e31c  2026-02-05 14:00:22.867890-06:00   \n",
+       "0  2026.03.24-18:25:55-03607e  2026-03-24 13:25:55.484936-05:00   \n",
+       "1  2026.03.24-18:20:18-8992ba  2026-03-24 13:20:18.221102-05:00   \n",
+       "2  2026.03.24-18:11:53-8992ba  2026-03-24 13:11:53.374532-05:00   \n",
+       "3  2026.03.24-17:51:33-8992ba  2026-03-24 12:51:33.144422-05:00   \n",
+       "4  2026.03.24-17:47:25-c80e5e  2026-03-24 12:47:25.169991-05:00   \n",
        "\n",
        "                       last_updated  n_trajectories  n_done_trajectories  \n",
-       "0  2026-02-05 14:24:37.524925-06:00               4                    4  \n",
-       "1  2026-02-05 14:20:07.585033-06:00               2                    2  \n",
-       "2  2026-02-05 14:17:55.001234-06:00               2                    2  \n",
-       "3  2026-02-05 14:13:44.568926-06:00               4                    1  \n",
-       "4  2026-02-05 14:01:13.190946-06:00               4                    0  "
+       "0  2026-03-24 13:34:43.160574-05:00              16                   16  \n",
+       "1  2026-03-24 13:24:23.428353-05:00              16                   16  \n",
+       "2  2026-03-24 13:12:42.391336-05:00              16                    4  \n",
+       "3  2026-03-24 12:52:17.679446-05:00              16                    3  \n",
+       "4  2026-03-24 12:49:43.698338-05:00               4                    4  "
       ]
      },
-     "execution_count": 5,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -184,7 +154,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "id": "60d2a8af-b656-4b82-b758-97d9ff5a4e57",
    "metadata": {},
    "outputs": [
@@ -218,7 +188,7 @@
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>221</th>\n",
+       "      <th>359</th>\n",
        "      <td>2025.10.31-18:21:40-451a69</td>\n",
        "      <td>2025-10-31 13:21:40.509101-05:00</td>\n",
        "      <td>2025-10-31 13:21:40.519187-05:00</td>\n",
@@ -226,7 +196,7 @@
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>222</th>\n",
+       "      <th>360</th>\n",
        "      <td>2025.10.31-18:18:57-451a69</td>\n",
        "      <td>2025-10-31 13:18:57.625552-05:00</td>\n",
        "      <td>2025-10-31 13:18:57.635521-05:00</td>\n",
@@ -234,7 +204,7 @@
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>223</th>\n",
+       "      <th>361</th>\n",
        "      <td>2025.10.31-18:18:16-451a69</td>\n",
        "      <td>2025-10-31 13:18:16.936863-05:00</td>\n",
        "      <td>2025-10-31 13:18:16.947114-05:00</td>\n",
@@ -242,7 +212,7 @@
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>224</th>\n",
+       "      <th>362</th>\n",
        "      <td>2025.10.31-18:17:34-451a69</td>\n",
        "      <td>2025-10-31 13:17:34.834340-05:00</td>\n",
        "      <td>2025-10-31 13:17:34.844208-05:00</td>\n",
@@ -250,7 +220,7 @@
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>225</th>\n",
+       "      <th>363</th>\n",
        "      <td>2025.10.31-18:17:13-451a69</td>\n",
        "      <td>2025-10-31 13:17:13.515018-05:00</td>\n",
        "      <td>2025-10-31 13:17:13.521163-05:00</td>\n",
@@ -263,21 +233,21 @@
       ],
       "text/plain": [
        "                         run_id                     first_created  \\\n",
-       "221  2025.10.31-18:21:40-451a69  2025-10-31 13:21:40.509101-05:00   \n",
-       "222  2025.10.31-18:18:57-451a69  2025-10-31 13:18:57.625552-05:00   \n",
-       "223  2025.10.31-18:18:16-451a69  2025-10-31 13:18:16.936863-05:00   \n",
-       "224  2025.10.31-18:17:34-451a69  2025-10-31 13:17:34.834340-05:00   \n",
-       "225  2025.10.31-18:17:13-451a69  2025-10-31 13:17:13.515018-05:00   \n",
+       "359  2025.10.31-18:21:40-451a69  2025-10-31 13:21:40.509101-05:00   \n",
+       "360  2025.10.31-18:18:57-451a69  2025-10-31 13:18:57.625552-05:00   \n",
+       "361  2025.10.31-18:18:16-451a69  2025-10-31 13:18:16.936863-05:00   \n",
+       "362  2025.10.31-18:17:34-451a69  2025-10-31 13:17:34.834340-05:00   \n",
+       "363  2025.10.31-18:17:13-451a69  2025-10-31 13:17:13.515018-05:00   \n",
        "\n",
        "                         last_updated  n_trajectories  n_done_trajectories  \n",
-       "221  2025-10-31 13:21:40.519187-05:00               2                    0  \n",
-       "222  2025-10-31 13:18:57.635521-05:00               2                    0  \n",
-       "223  2025-10-31 13:18:16.947114-05:00               2                    0  \n",
-       "224  2025-10-31 13:17:34.844208-05:00               2                    0  \n",
-       "225  2025-10-31 13:17:13.521163-05:00               2                    0  "
+       "359  2025-10-31 13:21:40.519187-05:00               2                    0  \n",
+       "360  2025-10-31 13:18:57.635521-05:00               2                    0  \n",
+       "361  2025-10-31 13:18:16.947114-05:00               2                    0  \n",
+       "362  2025-10-31 13:17:34.844208-05:00               2                    0  \n",
+       "363  2025-10-31 13:17:13.521163-05:00               2                    0  "
       ]
      },
-     "execution_count": 6,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -286,26 +256,36 @@
     "runs.tail()"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "467fb1a5-11a6-4990-944d-69e8c2cd439f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#last_run_id = good_run = '2026.02.26-18:47:48-cbece8'"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 7,
-   "id": "f1c340b4-7dc9-4ee2-bc9e-c81d26ad5129",
+   "id": "5da6f4a1-b2e4-431d-84bc-0b7eda1acc0f",
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "{'run_id': '2026.02.05-20:21:24-c0e31c',\n",
-       " 'n_trajectories': 4,\n",
-       " 'n_done': 4,\n",
+       "{'run_id': '2026.03.24-18:25:55-03607e',\n",
+       " 'n_trajectories': 16,\n",
+       " 'n_done': 16,\n",
        " 'n_active': 0,\n",
-       " 'total_chunks': 18,\n",
-       " 'total_passed_chunks': 8,\n",
-       " 'total_failed_chunks': 10,\n",
+       " 'total_chunks': 130,\n",
+       " 'total_passed_chunks': 64,\n",
+       " 'total_failed_chunks': 66,\n",
        " 'total_pending_chunks': 0,\n",
-       " 'total_training_frames': 50,\n",
-       " 'first_created': datetime.datetime(2026, 2, 5, 14, 21, 25, 16871, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=64800))),\n",
-       " 'last_updated': datetime.datetime(2026, 2, 5, 14, 24, 37, 524925, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=64800)))}"
+       " 'total_training_frames': 134,\n",
+       " 'first_created': datetime.datetime(2026, 3, 24, 13, 25, 55, 484936, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=68400))),\n",
+       " 'last_updated': datetime.datetime(2026, 3, 24, 13, 34, 43, 160574, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=68400)))}"
       ]
      },
      "execution_count": 7,
@@ -357,55 +337,187 @@
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td>0</td>\n",
-       "      <td>10</td>\n",
-       "      <td>2</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
        "      <td>TrajectoryStatus.COMPLETED</td>\n",
-       "      <td>2026-02-05 14:21:25.016871-06:00</td>\n",
-       "      <td>2026-02-05 14:23:41.302669-06:00</td>\n",
+       "      <td>2026-03-24 13:25:55.484936-05:00</td>\n",
+       "      <td>2026-03-24 13:34:43.160574-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>1</td>\n",
-       "      <td>10</td>\n",
-       "      <td>2</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
        "      <td>TrajectoryStatus.COMPLETED</td>\n",
-       "      <td>2026-02-05 14:21:25.108425-06:00</td>\n",
-       "      <td>2026-02-05 14:22:27.962408-06:00</td>\n",
+       "      <td>2026-03-24 13:25:55.582662-05:00</td>\n",
+       "      <td>2026-03-24 13:32:06.933508-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>2</td>\n",
-       "      <td>10</td>\n",
-       "      <td>2</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
        "      <td>TrajectoryStatus.COMPLETED</td>\n",
-       "      <td>2026-02-05 14:21:25.198387-06:00</td>\n",
-       "      <td>2026-02-05 14:24:37.524925-06:00</td>\n",
+       "      <td>2026-03-24 13:25:55.672487-05:00</td>\n",
+       "      <td>2026-03-24 13:33:46.339325-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>3</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:55.766757-05:00</td>\n",
+       "      <td>2026-03-24 13:33:26.811802-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>4</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:55.860735-05:00</td>\n",
+       "      <td>2026-03-24 13:33:47.806292-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>5</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:55.955970-05:00</td>\n",
+       "      <td>2026-03-24 13:33:30.704582-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>6</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.047957-05:00</td>\n",
+       "      <td>2026-03-24 13:32:56.622444-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>7</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.139428-05:00</td>\n",
+       "      <td>2026-03-24 13:29:32.569779-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>8</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.229928-05:00</td>\n",
+       "      <td>2026-03-24 13:32:15.566156-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>9</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.321930-05:00</td>\n",
+       "      <td>2026-03-24 13:29:42.980412-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
        "      <td>10</td>\n",
-       "      <td>2</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.412392-05:00</td>\n",
+       "      <td>2026-03-24 13:30:59.208321-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>11</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.506917-05:00</td>\n",
+       "      <td>2026-03-24 13:34:24.331417-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>12</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.601399-05:00</td>\n",
+       "      <td>2026-03-24 13:32:17.014580-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>13</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.693145-05:00</td>\n",
+       "      <td>2026-03-24 13:27:42.507302-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>14</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
+       "      <td>TrajectoryStatus.COMPLETED</td>\n",
+       "      <td>2026-03-24 13:25:56.788706-05:00</td>\n",
+       "      <td>2026-03-24 13:33:59.101814-05:00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>15</td>\n",
+       "      <td>40</td>\n",
+       "      <td>4</td>\n",
        "      <td>TrajectoryStatus.COMPLETED</td>\n",
-       "      <td>2026-02-05 14:21:25.286824-06:00</td>\n",
-       "      <td>2026-02-05 14:23:19.256584-06:00</td>\n",
+       "      <td>2026-03-24 13:25:56.883879-05:00</td>\n",
+       "      <td>2026-03-24 13:29:41.639440-05:00</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "   traj_id  target_length  chunks_completed                      status  \\\n",
-       "0        0             10                 2  TrajectoryStatus.COMPLETED   \n",
-       "1        1             10                 2  TrajectoryStatus.COMPLETED   \n",
-       "2        2             10                 2  TrajectoryStatus.COMPLETED   \n",
-       "3        3             10                 2  TrajectoryStatus.COMPLETED   \n",
+       "    traj_id  target_length  chunks_completed                      status  \\\n",
+       "0         0             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "1         1             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "2         2             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "3         3             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "4         4             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "5         5             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "6         6             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "7         7             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "8         8             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "9         9             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "10       10             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "11       11             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "12       12             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "13       13             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "14       14             40                 4  TrajectoryStatus.COMPLETED   \n",
+       "15       15             40                 4  TrajectoryStatus.COMPLETED   \n",
        "\n",
-       "                        created_at                       updated_at  \n",
-       "0 2026-02-05 14:21:25.016871-06:00 2026-02-05 14:23:41.302669-06:00  \n",
-       "1 2026-02-05 14:21:25.108425-06:00 2026-02-05 14:22:27.962408-06:00  \n",
-       "2 2026-02-05 14:21:25.198387-06:00 2026-02-05 14:24:37.524925-06:00  \n",
-       "3 2026-02-05 14:21:25.286824-06:00 2026-02-05 14:23:19.256584-06:00  "
+       "                         created_at                       updated_at  \n",
+       "0  2026-03-24 13:25:55.484936-05:00 2026-03-24 13:34:43.160574-05:00  \n",
+       "1  2026-03-24 13:25:55.582662-05:00 2026-03-24 13:32:06.933508-05:00  \n",
+       "2  2026-03-24 13:25:55.672487-05:00 2026-03-24 13:33:46.339325-05:00  \n",
+       "3  2026-03-24 13:25:55.766757-05:00 2026-03-24 13:33:26.811802-05:00  \n",
+       "4  2026-03-24 13:25:55.860735-05:00 2026-03-24 13:33:47.806292-05:00  \n",
+       "5  2026-03-24 13:25:55.955970-05:00 2026-03-24 13:33:30.704582-05:00  \n",
+       "6  2026-03-24 13:25:56.047957-05:00 2026-03-24 13:32:56.622444-05:00  \n",
+       "7  2026-03-24 13:25:56.139428-05:00 2026-03-24 13:29:32.569779-05:00  \n",
+       "8  2026-03-24 13:25:56.229928-05:00 2026-03-24 13:32:15.566156-05:00  \n",
+       "9  2026-03-24 13:25:56.321930-05:00 2026-03-24 13:29:42.980412-05:00  \n",
+       "10 2026-03-24 13:25:56.412392-05:00 2026-03-24 13:30:59.208321-05:00  \n",
+       "11 2026-03-24 13:25:56.506917-05:00 2026-03-24 13:34:24.331417-05:00  \n",
+       "12 2026-03-24 13:25:56.601399-05:00 2026-03-24 13:32:17.014580-05:00  \n",
+       "13 2026-03-24 13:25:56.693145-05:00 2026-03-24 13:27:42.507302-05:00  \n",
+       "14 2026-03-24 13:25:56.788706-05:00 2026-03-24 13:33:59.101814-05:00  \n",
+       "15 2026-03-24 13:25:56.883879-05:00 2026-03-24 13:29:41.639440-05:00  "
       ]
      },
      "execution_count": 8,
@@ -427,19 +539,21 @@
     {
      "data": {
       "text/plain": [
-       "{'run_id': '2026.02.05-20:21:24-c0e31c',\n",
+       "{'run_id': '2026.03.24-18:25:55-03607e',\n",
        " 'traj_id': 0,\n",
-       " 'target_length': 10,\n",
-       " 'chunks_completed': 2,\n",
-       " 'created_at': datetime.datetime(2026, 2, 5, 14, 21, 25, 16871, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=64800))),\n",
-       " 'updated_at': datetime.datetime(2026, 2, 5, 14, 23, 41, 302669, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=64800))),\n",
-       " 'n_chunk_attempts': 4,\n",
-       " 'n_unique_chunks': 2,\n",
-       " 'chunk_breakdown': {0: {'n_attempts': 2,\n",
+       " 'target_length': 40,\n",
+       " 'chunks_completed': 4,\n",
+       " 'created_at': datetime.datetime(2026, 3, 24, 13, 25, 55, 484936, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=68400))),\n",
+       " 'updated_at': datetime.datetime(2026, 3, 24, 13, 34, 43, 160574, tzinfo=datetime.timezone(datetime.timedelta(days=-1, seconds=68400))),\n",
+       " 'n_chunk_attempts': 15,\n",
+       " 'n_unique_chunks': 4,\n",
+       " 'chunk_breakdown': {0: {'n_attempts': 4,\n",
        "   'latest_status': 'PASSED',\n",
-       "   'latest_attempt_index': 1},\n",
-       "  1: {'n_attempts': 2, 'latest_status': 'PASSED', 'latest_attempt_index': 1}},\n",
-       " 'status_counts': {'PENDING': 0, 'PASSED': 2, 'FAILED': 2}}"
+       "   'latest_attempt_index': 3},\n",
+       "  1: {'n_attempts': 1, 'latest_status': 'PASSED', 'latest_attempt_index': 0},\n",
+       "  2: {'n_attempts': 1, 'latest_status': 'PASSED', 'latest_attempt_index': 0},\n",
+       "  3: {'n_attempts': 9, 'latest_status': 'PASSED', 'latest_attempt_index': 8}},\n",
+       " 'status_counts': {'PENDING': 0, 'PASSED': 4, 'FAILED': 11}}"
       ]
      },
      "execution_count": 9,
@@ -481,33 +595,41 @@
        "      <th></th>\n",
        "      <th>0</th>\n",
        "      <th>1</th>\n",
+       "      <th>2</th>\n",
+       "      <th>3</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>n_attempts</th>\n",
-       "      <td>2</td>\n",
-       "      <td>2</td>\n",
+       "      <td>4</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>9</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>latest_status</th>\n",
        "      <td>PASSED</td>\n",
        "      <td>PASSED</td>\n",
+       "      <td>PASSED</td>\n",
+       "      <td>PASSED</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>latest_attempt_index</th>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>8</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "                           0       1\n",
-       "n_attempts                 2       2\n",
-       "latest_status         PASSED  PASSED\n",
-       "latest_attempt_index       1       1"
+       "                           0       1       2       3\n",
+       "n_attempts                 4       1       1       9\n",
+       "latest_status         PASSED  PASSED  PASSED  PASSED\n",
+       "latest_attempt_index       3       0       0       8"
       ]
      },
      "execution_count": 10,
@@ -591,217 +713,155 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
-       "      <td>5</td>\n",
+       "      <td>10</td>\n",
        "      <td>FAILED</td>\n",
        "      <td>0</td>\n",
-       "      <td>2026-02-05 14:22:11.141181-06:00</td>\n",
-       "      <td>2026-02-05 14:22:12.769573-06:00</td>\n",
+       "      <td>2026-03-24 13:25:57.559453-05:00</td>\n",
+       "      <td>2026-03-24 13:26:45.434941-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>PASSED</td>\n",
+       "      <td>10</td>\n",
+       "      <td>FAILED</td>\n",
        "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:08.032971-06:00</td>\n",
-       "      <td>2026-02-05 14:23:09.063405-06:00</td>\n",
+       "      <td>2026-03-24 13:28:16.990763-05:00</td>\n",
+       "      <td>2026-03-24 13:28:59.450223-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>0</td>\n",
-       "      <td>1</td>\n",
        "      <td>0</td>\n",
-       "      <td>5</td>\n",
+       "      <td>2</td>\n",
+       "      <td>10</td>\n",
        "      <td>FAILED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:20.500798-06:00</td>\n",
-       "      <td>2026-02-05 14:23:21.786672-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>PASSED</td>\n",
        "      <td>2</td>\n",
-       "      <td>2026-02-05 14:23:40.443672-06:00</td>\n",
-       "      <td>2026-02-05 14:23:41.302669-06:00</td>\n",
+       "      <td>2026-03-24 13:30:22.527519-05:00</td>\n",
+       "      <td>2026-03-24 13:30:55.740054-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>FAILED</td>\n",
+       "      <th>3</th>\n",
        "      <td>0</td>\n",
-       "      <td>2026-02-05 14:22:11.141181-06:00</td>\n",
-       "      <td>2026-02-05 14:22:12.769573-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>1</td>\n",
        "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "      <td>10</td>\n",
        "      <td>PASSED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:08.032971-06:00</td>\n",
-       "      <td>2026-02-05 14:23:09.063405-06:00</td>\n",
+       "      <td>4</td>\n",
+       "      <td>2026-03-24 13:31:20.628141-05:00</td>\n",
+       "      <td>2026-03-24 13:31:48.009820-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
+       "      <th>4</th>\n",
        "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>FAILED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:20.500798-06:00</td>\n",
-       "      <td>2026-02-05 14:23:21.786672-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>1</td>\n",
        "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>PASSED</td>\n",
-       "      <td>2</td>\n",
-       "      <td>2026-02-05 14:23:40.443672-06:00</td>\n",
-       "      <td>2026-02-05 14:23:41.302669-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>FAILED</td>\n",
        "      <td>0</td>\n",
-       "      <td>2026-02-05 14:22:11.141181-06:00</td>\n",
-       "      <td>2026-02-05 14:22:12.769573-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
+       "      <td>10</td>\n",
        "      <td>PASSED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:08.032971-06:00</td>\n",
-       "      <td>2026-02-05 14:23:09.063405-06:00</td>\n",
+       "      <td>4</td>\n",
+       "      <td>2026-03-24 13:31:48.643006-05:00</td>\n",
+       "      <td>2026-03-24 13:32:01.039034-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>FAILED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:20.500798-06:00</td>\n",
-       "      <td>2026-02-05 14:23:21.786672-06:00</td>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>PASSED</td>\n",
-       "      <td>2</td>\n",
-       "      <td>2026-02-05 14:23:40.443672-06:00</td>\n",
-       "      <td>2026-02-05 14:23:41.302669-06:00</td>\n",
+       "      <th>10</th>\n",
+       "      <td>15</td>\n",
+       "      <td>3</td>\n",
+       "      <td>4</td>\n",
+       "      <td>10</td>\n",
+       "      <td>FAILED</td>\n",
+       "      <td>13</td>\n",
+       "      <td>2026-03-24 13:34:06.130235-05:00</td>\n",
+       "      <td>2026-03-24 13:34:12.841984-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>0</th>\n",
+       "      <th>11</th>\n",
+       "      <td>15</td>\n",
        "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
        "      <td>5</td>\n",
+       "      <td>10</td>\n",
        "      <td>FAILED</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2026-02-05 14:22:11.141181-06:00</td>\n",
-       "      <td>2026-02-05 14:22:12.769573-06:00</td>\n",
+       "      <td>14</td>\n",
+       "      <td>2026-03-24 13:34:13.434618-05:00</td>\n",
+       "      <td>2026-03-24 13:34:18.915203-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>1</th>\n",
+       "      <th>12</th>\n",
+       "      <td>15</td>\n",
        "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>PASSED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:08.032971-06:00</td>\n",
-       "      <td>2026-02-05 14:23:09.063405-06:00</td>\n",
+       "      <td>6</td>\n",
+       "      <td>10</td>\n",
+       "      <td>FAILED</td>\n",
+       "      <td>15</td>\n",
+       "      <td>2026-03-24 13:34:21.308209-05:00</td>\n",
+       "      <td>2026-03-24 13:34:28.284012-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2</th>\n",
+       "      <th>13</th>\n",
+       "      <td>15</td>\n",
        "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
+       "      <td>7</td>\n",
+       "      <td>10</td>\n",
        "      <td>FAILED</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2026-02-05 14:23:20.500798-06:00</td>\n",
-       "      <td>2026-02-05 14:23:21.786672-06:00</td>\n",
+       "      <td>16</td>\n",
+       "      <td>2026-03-24 13:34:29.457115-05:00</td>\n",
+       "      <td>2026-03-24 13:34:34.904288-05:00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>3</th>\n",
+       "      <th>14</th>\n",
+       "      <td>15</td>\n",
        "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5</td>\n",
+       "      <td>8</td>\n",
+       "      <td>10</td>\n",
        "      <td>PASSED</td>\n",
-       "      <td>2</td>\n",
-       "      <td>2026-02-05 14:23:40.443672-06:00</td>\n",
-       "      <td>2026-02-05 14:23:41.302669-06:00</td>\n",
+       "      <td>17</td>\n",
+       "      <td>2026-03-24 13:34:36.370942-05:00</td>\n",
+       "      <td>2026-03-24 13:34:43.160574-05:00</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
+       "<p>240 rows × 8 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "   traj_id  chunk_id  attempt_index  n_frames audit_status  model_version  \\\n",
-       "0        0         0              0         5       FAILED              0   \n",
-       "1        0         0              1         5       PASSED              1   \n",
-       "2        0         1              0         5       FAILED              1   \n",
-       "3        0         1              1         5       PASSED              2   \n",
-       "0        1         0              0         5       FAILED              0   \n",
-       "1        1         0              1         5       PASSED              1   \n",
-       "2        1         1              0         5       FAILED              1   \n",
-       "3        1         1              1         5       PASSED              2   \n",
-       "0        2         0              0         5       FAILED              0   \n",
-       "1        2         0              1         5       PASSED              1   \n",
-       "2        2         1              0         5       FAILED              1   \n",
-       "3        2         1              1         5       PASSED              2   \n",
-       "0        3         0              0         5       FAILED              0   \n",
-       "1        3         0              1         5       PASSED              1   \n",
-       "2        3         1              0         5       FAILED              1   \n",
-       "3        3         1              1         5       PASSED              2   \n",
+       "    traj_id  chunk_id  attempt_index  n_frames audit_status  model_version  \\\n",
+       "0         0         0              0        10       FAILED              0   \n",
+       "1         0         0              1        10       FAILED              1   \n",
+       "2         0         0              2        10       FAILED              2   \n",
+       "3         0         0              3        10       PASSED              4   \n",
+       "4         0         1              0        10       PASSED              4   \n",
+       "..      ...       ...            ...       ...          ...            ...   \n",
+       "10       15         3              4        10       FAILED             13   \n",
+       "11       15         3              5        10       FAILED             14   \n",
+       "12       15         3              6        10       FAILED             15   \n",
+       "13       15         3              7        10       FAILED             16   \n",
+       "14       15         3              8        10       PASSED             17   \n",
        "\n",
-       "                        created_at                       updated_at  \n",
-       "0 2026-02-05 14:22:11.141181-06:00 2026-02-05 14:22:12.769573-06:00  \n",
-       "1 2026-02-05 14:23:08.032971-06:00 2026-02-05 14:23:09.063405-06:00  \n",
-       "2 2026-02-05 14:23:20.500798-06:00 2026-02-05 14:23:21.786672-06:00  \n",
-       "3 2026-02-05 14:23:40.443672-06:00 2026-02-05 14:23:41.302669-06:00  \n",
-       "0 2026-02-05 14:22:11.141181-06:00 2026-02-05 14:22:12.769573-06:00  \n",
-       "1 2026-02-05 14:23:08.032971-06:00 2026-02-05 14:23:09.063405-06:00  \n",
-       "2 2026-02-05 14:23:20.500798-06:00 2026-02-05 14:23:21.786672-06:00  \n",
-       "3 2026-02-05 14:23:40.443672-06:00 2026-02-05 14:23:41.302669-06:00  \n",
-       "0 2026-02-05 14:22:11.141181-06:00 2026-02-05 14:22:12.769573-06:00  \n",
-       "1 2026-02-05 14:23:08.032971-06:00 2026-02-05 14:23:09.063405-06:00  \n",
-       "2 2026-02-05 14:23:20.500798-06:00 2026-02-05 14:23:21.786672-06:00  \n",
-       "3 2026-02-05 14:23:40.443672-06:00 2026-02-05 14:23:41.302669-06:00  \n",
-       "0 2026-02-05 14:22:11.141181-06:00 2026-02-05 14:22:12.769573-06:00  \n",
-       "1 2026-02-05 14:23:08.032971-06:00 2026-02-05 14:23:09.063405-06:00  \n",
-       "2 2026-02-05 14:23:20.500798-06:00 2026-02-05 14:23:21.786672-06:00  \n",
-       "3 2026-02-05 14:23:40.443672-06:00 2026-02-05 14:23:41.302669-06:00  "
+       "                         created_at                       updated_at  \n",
+       "0  2026-03-24 13:25:57.559453-05:00 2026-03-24 13:26:45.434941-05:00  \n",
+       "1  2026-03-24 13:28:16.990763-05:00 2026-03-24 13:28:59.450223-05:00  \n",
+       "2  2026-03-24 13:30:22.527519-05:00 2026-03-24 13:30:55.740054-05:00  \n",
+       "3  2026-03-24 13:31:20.628141-05:00 2026-03-24 13:31:48.009820-05:00  \n",
+       "4  2026-03-24 13:31:48.643006-05:00 2026-03-24 13:32:01.039034-05:00  \n",
+       "..                              ...                              ...  \n",
+       "10 2026-03-24 13:34:06.130235-05:00 2026-03-24 13:34:12.841984-05:00  \n",
+       "11 2026-03-24 13:34:13.434618-05:00 2026-03-24 13:34:18.915203-05:00  \n",
+       "12 2026-03-24 13:34:21.308209-05:00 2026-03-24 13:34:28.284012-05:00  \n",
+       "13 2026-03-24 13:34:29.457115-05:00 2026-03-24 13:34:34.904288-05:00  \n",
+       "14 2026-03-24 13:34:36.370942-05:00 2026-03-24 13:34:43.160574-05:00  \n",
+       "\n",
+       "[240 rows x 8 columns]"
       ]
      },
      "execution_count": 13,
@@ -820,15 +880,23 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": 14,
-   "id": "539119da-ff93-4f07-94b5-a96b13fa184e",
+   "cell_type": "markdown",
+   "id": "2a654ce0-80d0-45bf-bd79-4246285941ea",
    "metadata": {},
-   "outputs": [
+   "source": [
+    "### These statuses are getting set at goofy times. the log says that these passed or failed, but theyre pending in the db..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "539119da-ff93-4f07-94b5-a96b13fa184e",
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "(8, 8)"
+       "(64, 8)"
       ]
      },
      "execution_count": 14,
@@ -850,7 +918,7 @@
     {
      "data": {
       "text/plain": [
-       "8"
+       "64"
       ]
      },
      "execution_count": 15,
@@ -870,7 +938,127 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHFCAYAAADv8c1wAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABWg0lEQVR4nO3deVzU1f7H8dewiwIqKIsrmhviBqShqW3icq+peYsWl+peb3RNRW43bTXbzHtv/VrcsmurpVZmaplXrNRM1ETEDXcUF5BwAVzYZr6/P0xuBOKAyAzD+/l4zKM485nvfGZK5u35njlfk2EYBiIiIiJSLidbNyAiIiJSEyg0iYiIiFhBoUlERETECgpNIiIiIlZQaBIRERGxgkKTiIiIiBUUmkRERESsoNAkIiIiYgWFJhERERErKDSJiMMwmUxW3dasWXNNz/P8889jMpmsqs3MzOTBBx/Ez88PT09PIiMj+e67767p+UXENky6jIqIOIqNGzeW+PnFF1/khx9+4Pvvvy8xHhISgre3d6Wf59ixYxw7doybbrqp3Lr8/HwiIiI4e/Ysr776Ko0bN2bmzJl88803rF69mr59+1a6BxGpfgpNIuKwHnzwQb744gvOnTtXbt2FCxfw9PSs8uefNWsWY8eOZcOGDURGRgJQVFREly5dqFevHps2bary5xSR60en50SkVrnlllsIDQ1l3bp19OzZE09PTx5++GEAFi1aRFRUFIGBgdSpU4cOHTowefJkzp8/X+IY1p6eW7JkCe3atSsOTAAuLi6MGDGCzZs3c/z48ap9cSJyXbnYugERkeqWnp7OiBEjeOKJJ3jllVdwcrr098f9+/czaNAgYmNjqVu3Lnv27GH69Ols3ry51Ck+a+zcuZPevXuXGu/cuTMAu3btokmTJtf2YkSk2ig0iUitc/r0aT7//HNuu+22EuPPPPNM8b8bhkGvXr3o0KEDffv2Zfv27cVhx1qnTp2iYcOGpcYvj506daoS3YuIrej0nIjUOg0aNCgVmAAOHTrE/fffT0BAAM7Ozri6uhYv1k5JSanUc5V3Gs/ab+CJiH3QTJOI1DqBgYGlxs6dO0fv3r3x8PDgpZdeom3btnh6enL06FHuuusuLl68WOHn8fX1LXM26fTp0wBlzkKJiP1SaBKRWqesGZ7vv/+eEydOsGbNmhJbAZw9e7bSz9OpUyd27NhRavzyWGhoaKWPLSLVT6fnRET4X5Byd3cvMf7OO+9U+pjDhg1jz549JbYWKCoqYv78+fTo0YOgoKBKH1tEqp9mmkREgJ49e9KgQQNiYmKYMmUKrq6ufPLJJyQnJ1f6mA8//DAzZ87k7rvvLt7cctasWezdu5fVq1dXYfciUh000yQiwqX1R9988w2enp6MGDGChx9+mHr16rFo0aJKH9Pd3Z3vvvuOW2+9lXHjxjF48GDS09P59ttvtRu4SA2kHcFFRCpo4sSJfPzxx2RlZdm6FRGpRjo9JyJipczMTBISEvjyyy9L7PItIrWDTs+JiFhpxYoVPPDAA7Rp04Y333zT1u2ISDXT6TkRERERK2imSURERMQKCk0iIiIiVlBoEhEREbGCvj1XRSwWCydOnMDLy0sX4RQREakhDMMgNzeXoKAgnJzKn0tSaKoiJ06coFmzZrZuQ0RERCrh6NGjNG3atNwahaYq4uXlBVx60729vW3cjYiIiFgjJyeHZs2aFX+Ol0ehqYpcPiXn7e2t0CQiIlLDWLO0RgvBRURERKyg0CQiIiJiBYUmERERESsoNImIiIhYQaFJRERExAoKTSIiIiJWUGgSERERsYJCk4iIiIgVFJpERERErKAdwUVERMSumS0Gm1NPk5mbR2MvD7oHN8TZ6eo7eFc1hSYRERGxWyt3pjN1+W7Ss/OKxwJ9PJgyOIQBoYHV2otOz4mIiIhdWrkznUfnby0RmAAysvN4dP5WVu5Mr9Z+FJpERETE7pgtBlOX78Yo477LY1OX78ZsKavi+lBoEhEREbuzOfV0qRmm3zKA9Ow8Nqeerrae7C40zZo1i+DgYDw8PAgPD+fHH38st37t2rWEh4fj4eFBq1atmDNnTqmaxYsXExISgru7OyEhISxZsqTE/c8//zwmk6nELSAgoEpfl4iIiFgvM/fKgakydVXBrkLTokWLiI2N5emnnyYpKYnevXszcOBA0tLSyqxPTU1l0KBB9O7dm6SkJJ566inGjx/P4sWLi2sSEhKIjo5m5MiRJCcnM3LkSO655x42bdpU4lgdO3YkPT29+LZjx47r+lpFRETkyhp7eVRpXVUwGYZRfScDr6JHjx6EhYUxe/bs4rEOHTowdOhQpk2bVqp+0qRJLFu2jJSUlOKxmJgYkpOTSUhIACA6OpqcnBy+/fbb4poBAwbQoEEDFixYAFyaafrqq6/Ytm1bpXvPycnBx8eH7OxsvL29K30cERERgf/8eIiXvkm54v0mIMDHg/WTbrum7Qcq8vltNzNNBQUFJCYmEhUVVWI8KiqKDRs2lPmYhISEUvX9+/dny5YtFBYWllvz+2Pu37+foKAggoODuffeezl06NC1viQRERGphA9+Sr1qYAKYMjikWvdrspvQlJWVhdlsxt/fv8S4v78/GRkZZT4mIyOjzPqioiKysrLKrfntMXv06MFHH33Ef//7X959910yMjLo2bMnp06dumK/+fn55OTklLiJiIjItflww2GeX74bgL/d0prZD3Qj0KfkKbgAHw9mjwir9n2a7G5zS5OpZGI0DKPU2NXqfz9+tWMOHDiw+N87depEZGQkrVu35sMPPyQuLq7M5502bRpTp069yqsRERERa32UcJgpy3YB8OgtrflH/3aYTCaiOgZqR/Df8vPzw9nZudSsUmZmZqmZossCAgLKrHdxccHX17fcmisdE6Bu3bp06tSJ/fv3X7HmySefLBGocnJyaNas2RXrRURE5Mo+TjjMc0svBaaYvq154tfABODsZCKyta8t2wPs6PScm5sb4eHhxMfHlxiPj4+nZ8+eZT4mMjKyVP2qVauIiIjA1dW13JorHRMunXpLSUkhMPDK037u7u54e3uXuImIiEjFfbzxCM/+Gpge6dOKSQPalXuWyVbsJjQBxMXF8Z///If33nuPlJQUJk6cSFpaGjExMcCl2Z1Ro0YV18fExHDkyBHi4uJISUnhvffeY968eTz++OPFNRMmTGDVqlVMnz6dPXv2MH36dFavXk1sbGxxzeOPP87atWtJTU1l06ZN/OlPfyInJ4fRo0dX22sXERGpjT7ZdIRnv9oJwF/7tGLywPZ2GZjAjk7PwaXtAU6dOsULL7xAeno6oaGhrFixghYtWgCQnp5eYs+m4OBgVqxYwcSJE5k5cyZBQUG89dZbDB8+vLimZ8+eLFy4kGeeeYZnn32W1q1bs2jRInr06FFcc+zYMe677z6ysrJo1KgRN910Exs3bix+XhEREal6n25K4+kllwLTmN7BPGnHgQnsbJ+mmkz7NImIiFhvweY0nvzy0kbSf7k5mKf/0MEmgalG7tMkIiIitcPC3wSmh3vZLjBVlEKTiIiIVJtFP6cx+dfA9FCvljz7x5oRmEChSURERKrJZz8fLQ5MD/ZsyXN/DKkxgQkUmkRERKQafLblKJO+3I5hXApMUwbXrMAECk0iIiJynX2ReIxJiy8FptGRLWpkYAKFJhEREbmOvkg8xj++SMYwYFRkC56/s2ONDEyg0CQiIiLXyZdb/xeYRtzUnKk1ODCBQpOIiIhcB0uSjvH3zy8Fpgd6NOeFO0NrdGAChSYRERGpYl8lHefvn10KTPf3aM6LQ0JxcqrZgQkUmkRERKQKLd12nLjPtmEx4L7uzXnJQQITKDSJiIhIFVm67TgTF10KTPfe2IyXhzpOYAKFJhEREakCy5JPlAhMrwzr5FCBCRSaRERE5BotTz5B7MIkLAZERzhmYAKFJhEREbkGX28/QeyvM0x3hzdl2l2OGZhAoUlEREQq6Zvt6UxYuA2zxeBP4U2ZPryzwwYmUGgSERGRSlixI53xC5MwWwyGhzl+YAKFJhEREamgb3ekM27BpcB0V1gT/vmnzjg7eGAChSYRERGpgJU7fxOYujXhX3/qUisCEyg0iYiIiJVW7szgsU+TKLIYDOvWhH/dXXsCEyg0iYiIiBX+uyuDxz7dSpHFYGjXIP5dywITKDSJiIjIVazalcHYTy4FpiFdg3jtnq61LjCBQpOIiIiUI373Scb+OsM0uEsQr9XCGabLFJpERESkTKt3n+RvnyRSaL4UmP7vni64ONfe6FB7X7mIiIhc0XcpJ3n018D0x86BtT4wgUKTiIiI/M73e07y6PytFJoN/tApkDeiu9b6wAQKTSIiIvIbP+zJJObjrRSYLQzqFMAb9yowXaZ3QURERAD4YW8mj3ycSIHZwsDQAN68txuuCkzF9E6IiIgIa34TmAZ0DOCt+xSYfk/vhoiISC23dt8v/PXjRAqKLPTv6M/b9yswlcXF1g2IiIhI9TFbDDanniYzN4/GXh7kF5mLA1NUiD9v3xemwHQFCk0iIiK1xMqd6Uxdvpv07LxS9/UL8WfG/WG4uSgwXYlCk4iISC2wcmc6j87finGF+4d0CVRgugq9OyIiIg7ObDGYunz3FQMTwMsr9mC2lFchCk0iIiIObnPq6TJPyf1WenYem1NPV1NHNZNCk4iIiIPLzC0/MFW0rrZSaBIREXFwjb08qrSuttJCcBEREQeXkX2x3PtNQICPB92DG1ZPQzWUZppEREQc2Gc/HyXu8+Qr3m/69Z9TBofg7GS6Yp0oNImIiDisjxMO88Ti7RgGPNCjObMe6EagT8lTcAE+HsweEcaA0EAbdVlz6PSciIiIA/rPj4d46ZsUAB7uFcyzf+yAyWSif8fAEjuCdw9uqBkmKyk0iYiIOJi3v9vPa/H7ABh7a2sej2qHyXQpGDk7mYhs7WvL9moshSYREREHYRgG/161l5k/HATg7/3aMu72NjbuynEoNImIiDgAwzB46ZsU5q1PBeDpQR0Y06eVjbtyLApNIiIiNZzFYvDs0p18sikNgBeGdGRUZEvbNuWAFJpERERqMLPFYNLi7XyReAyTCabf1Zl7bmxm67YckkKTiIhIDVVotjBx0Ta+3p6Os5OJ1+/pwpCuTWzdlsNSaBIREamB8ovMjPs0iVW7T+LqbOKte7sxsJP2WrqeFJpERERqmLxCMzHzE1mz9xfcXJyYMyKM29r727oth6fQJCIiUoNcKCjiLx9uYcPBU3i4OvGfUTdycxs/W7dVKyg0iYiI1BC5eYU89P7PbDlyhrpuzrz/UHddZLcaKTSJiIjUAGcvFDD6vc0kH8vG28OFDx/uTrfmDWzdVq2i0CQiImLnTp3LZ8S8zaSk59DA05WP/9yD0CY+tm6r1lFoEhERsWOZOXnc/59NHMg8h189dz4d04O2/l62bqtWUmgSERGxU8fPXuSBdzdy+NQFAn08+OQvPWjVqJ6t26q1FJpERETsUNqpC9z37kaOn71I0wZ1WDDmJpo19LR1W7WaQpOIiIidOZB5jgf+s5GTOfkE+9Xlk7/0IKh+HVu3Ves52bqB35s1axbBwcF4eHgQHh7Ojz/+WG792rVrCQ8Px8PDg1atWjFnzpxSNYsXLyYkJAR3d3dCQkJYsmTJFY83bdo0TCYTsbGx1/pSREREKmxPRg73zk3gZE4+bf3rseiRmxSY7IRdhaZFixYRGxvL008/TVJSEr1792bgwIGkpaWVWZ+amsqgQYPo3bs3SUlJPPXUU4wfP57FixcX1yQkJBAdHc3IkSNJTk5m5MiR3HPPPWzatKnU8X7++Wfmzp1L586dr9trFBERuZKdx7O5d+5Gss4VEBLozcK/RtLYy8PWbcmvTIZhGLZu4rIePXoQFhbG7Nmzi8c6dOjA0KFDmTZtWqn6SZMmsWzZMlJSUorHYmJiSE5OJiEhAYDo6GhycnL49ttvi2sGDBhAgwYNWLBgQfHYuXPnCAsLY9asWbz00kt07dqVN954w+rec3Jy8PHxITs7G29v74q8bBERERKPnOHB9zeTm1dEl2b1+eih7vh4utq6LYdXkc9vu5lpKigoIDExkaioqBLjUVFRbNiwoczHJCQklKrv378/W7ZsobCwsNya3x9z7Nix/OEPf+COO+6wqt/8/HxycnJK3ERERCpj46FTjJq3idy8Im5s2YD5f1Zgskd2E5qysrIwm834+5e84KC/vz8ZGRllPiYjI6PM+qKiIrKyssqt+e0xFy5cyNatW8uczbqSadOm4ePjU3xr1qyZ1Y8VERG5bN2+X3jw/c2cLzDT6wZfPny4O14eCkz2yG5C02Umk6nEz4ZhlBq7Wv3vx8s75tGjR5kwYQLz58/Hw8P688ZPPvkk2dnZxbejR49a/VgRERGA1btP8pcPt5BXaOG29o2ZN/pGPN30xXZ7ZTf/Zfz8/HB2di41q5SZmVlqpuiygICAMutdXFzw9fUtt+byMRMTE8nMzCQ8PLz4frPZzLp165gxYwb5+fk4OzuXem53d3fc3d0r/kJFRESAb7anM2FhEkUWgwEdA3jrvm64udjdXIb8ht3813FzcyM8PJz4+PgS4/Hx8fTs2bPMx0RGRpaqX7VqFREREbi6upZbc/mYt99+Ozt27GDbtm3Ft4iICB544AG2bdtWZmASERG5FkuSjjFuwVaKLAZDugYx434FpprAbmaaAOLi4hg5ciQRERFERkYyd+5c0tLSiImJAS6dEjt+/DgfffQRcOmbcjNmzCAuLo4xY8aQkJDAvHnzSnwrbsKECfTp04fp06czZMgQli5dyurVq1m/fj0AXl5ehIaGluijbt26+Pr6lhoXERG5Vgs3p/Hkkh0YBtwT0ZRpd3XG2enKy1DEfthVaIqOjubUqVO88MILpKenExoayooVK2jRogUA6enpJfZsCg4OZsWKFUycOJGZM2cSFBTEW2+9xfDhw4trevbsycKFC3nmmWd49tlnad26NYsWLaJHjx7V/vpERKR2++CnVJ5fvhuAkTe1YOqdHXFSYKox7GqfpppM+zSJiEh55qw9yKvf7gFgTO9gnhrUodwvOkn1qMjnt13NNImIiDgawzB487v9vLF6PwDjbruBuH5tFZhqIIUmERGR68QwDKav3MuctQcB+Ef/doy99QYbdyWVpdAkIiJyHRiGwdTlu/lgw2EAnv1jCH++Odi2Tck1UWgSERGpYhaLwdNf7WDB5ksbH780NJQRN7WwcVdyrRSaREREqlCR2cITX2zny6TjOJlg+vDO3B2hS205AoUmERGRKlJothC7cBvf7EjH2cnE/0V35c4uQbZuS6qIQpOIiEglmC0Gm1NPk5mbR2MvD7o082H8gm2sTjmJq7OJGfeH0b9jgK3blCqk0CQiIlJBK3emM3X5btKz84rH3FycKCiy4O7ixJyR4dzarrENO5TrQaFJRESkAlbuTOfR+Vv5/c7QBUUWAB69pbUCk4PS1QFFRESsZLZc2kagvEtpLPr5KGaLLrbhiBSaRERErLQ59XSJU3JlSc/OY3Pq6WrqSKqTQpOIiIiVMnPLD0wVrZOaRaFJRETESo29PKq0TmoWLQQXERGx0p6MnHLvNwEBPh50D25YPQ1JtVJoEhERuQqzxeDlb1J476fUK9aYfv3nlMEhODuZrlgnNZdOz4mIiJTjYoGZR+cnFgemf/Rvx+wHuhHoU/IUXICPB7NHhDEgNNAWbUo10EyTiIjIFWTm5jHmwy0kH8vGzcWJ1+7uwuBfL4sS1TGwxI7g3YMbaobJwSk0iYiIlGH/yVwefP9njp+9SANPV94dFUFEy/+tVXJ2MhHZ2teGHUp1U2gSERH5nQ0HsnhkfiK5eUW09PXk/Ye6E+xX19ZtiY0pNImIiPzGF4nHmLx4O0UWg4gWDZg7KoKGdd1s3ZbYAYUmERERwDAM/i9+H299fwCAwV2C+NefOuPh6mzjzsReKDSJiEitl19kZtIX2/lq2wkAxt7amr/3a4eTFnbLbyg0iYhIrXb2QgF//TiRzamncXYy8cqwUKJvbG7rtsQOKTSJiEitdeTUeR764GcO/XIeL3cXZo0Io3ebRrZuS+yUQpOIiNRKiUfOMOajLZw+X0CQjwfvP9SddgFetm5L7JhCk4iI1DrfbE8n7rNt5BdZCG3izXujb6Sxty6yK+VTaBIRkVrDMAzeWXeIV7/dA8AdHRrz5r3dqOuuj0O5Ov1fIiIitUKR2cJzy3bx6aY0AB7s2ZJn/6iL64r1FJpERMThncsvYuwnW1m77xdMJnj2DyE8fHOwrduSGkahSUREHFp69kUeev9n9mTk4uHqxFv3diOqY4Ct25IaSKFJREQc1q4T2Tz8wc+czMnHr54780ZH0KVZfVu3JTWUQpOIiDikH/Zk8tinWzlfYKZN43q89+CNNGvoaeu2pAZTaBIREYfz8cYjTFm6E4sBvW7wZdYD4fjUcbV1W1LDKTSJiIjDsFgMXl25h7nrDgFwd3hTXh7WCTcXJxt3Jo5AoUlERBxCXqGZiYu28e3ODAAej2rL2FtvwGTSlgJSNRSaRESkxss6l89fPtzCtqNncXN24l93d2ZI1ya2bkscjEKTiIjUaAcyz/HQB5s5evoi9T1deWdEOD1a+dq6LXFACk0iIlJjbTx0ir9+tIWcvCJa+Hry/oM30qpRPVu3JQ5KoUlERGqkL7ceY9Li7RSaDcKa1+fdURH41nO3dVviwCoUmpYtW1bhJ+jXrx916tSp8ONERETKYhgGb363nzdW7wfgD50Cee2eLni4Otu4M3F0FQpNQ4cOrdDBTSYT+/fvp1WrVhV6nIiISFkKiixM/nI7X249DkBM39Y80b8dTrrorlSDCp+ey8jIoHHjxlbVenl5VbghERGRsmRfKCRmfiIJh07h7GTipaGh3Ne9ua3bklqkQqFp9OjRFTrVNmLECLy9vSvclIiIyG8dPX2BB9/fzMFfzlPP3YWZD4TRt20jW7cltYzJMAzD1k04gpycHHx8fMjOzlZQFBGpQklpZxjz0RayzhUQ6OPBew/eSIdA/Z6VqlGRz299e05EROzWyp3pTFi4jfwiCyGB3rz34I0E+HjYui2ppSoUmoKDgyu1HX1sbCzjx4+v8ONERKR2MFsMNqeeJjM3j8ZeHtzYsgEfbDjMyytSMAy4tV0j3r4/jHru+ru+2E6F/u/74IMPKvUkLVu2rNTjRETE8a3cmc7U5btJz84rHvN0c+ZCgRmAkTe1YMrgEFycddFdsa0Khaa+ffterz5ERKQWWrkznUfnb+X3i2svB6Y/hTXhhSEdddFdsQtVFtvz8/Or6lAiIlILmC0GU5fvLhWYfuung6ew6OtKYieqLDT17Nmz1Ni+ffuq6vAiIuJgNqeeLnFKrizp2XlsTj1dTR2JlO+aV9R9/fXX7Nmzh/Pnz3PixAmCgoKK77v77rtJTk6+1qcQEREHlJlbfmCqaJ3I9XbNoaljx46kpaWRmZnJfffdx9GjR2natClBQUE4O+s6QCIiUraGdd2sqmvspS0GxD5cc2gKDg7mb3/7G6GhofTp0weA48ePk5qaSmho6DU3KCIijufUuXxmfH+g3BoTEODjQffghtXTlMhVVCo0FRYWMmDAAOrWrcvSpUsxmUzFgQmgSZMmNGnSpMqaFBERx7HzeDaPfJzI8bMX8XB1Iq/QgglKLAi//F25KYNDcNbFeMVOVGohuKurK1u3biUtLU1fAxUREast3XacP83ZwPGzFwn2q8vX425mzoiwUrt8B/h4MHtEGANCA23UqUhplf723LBhwzhw4ADHjh2ryn6YNWsWwcHBeHh4EB4ezo8//lhu/dq1awkPD8fDw4NWrVoxZ86cUjWLFy8mJCQEd3d3QkJCWLJkSYn7Z8+eTefOnfH29sbb25vIyEi+/fbbKn1dIiK1mdliMG1FChMWbiOv0MIt7Rrx1dhe3NDYiwGhgayfdBsLxtzEm/d2ZcGYm1g/6TYFJrE7lV7T5OfnR1FREWFhYQwfPpyAgIASs07PPfdchY+5aNEiYmNjmTVrFr169eKdd95h4MCB7N69m+bNm5eqT01NZdCgQYwZM4b58+fz008/8be//Y1GjRoxfPhwABISEoiOjubFF19k2LBhLFmyhHvuuYf169fTo0cPAJo2bcqrr77KDTfcAMCHH37IkCFDSEpKomPHjpV5e0RE5FfZFwp5bMFWftyfBcCjt7Tm8ah2JU67OTuZiGzta6sWRaxiMgyjUtuGOTk5YTKZMAyjzFN0ZrO5wsfs0aMHYWFhzJ49u3isQ4cODB06lGnTppWqnzRpEsuWLSMlJaV4LCYmhuTkZBISEgCIjo4mJyenxMzRgAEDaNCgAQsWLLhiLw0bNuRf//oXf/7zn63qvSJXSRYRqS32ncxlzEdbOHLqAnVcnfnX3Z35Y+egqz9QpJpU5PO70jNNzZs3r9L1TAUFBSQmJjJ58uQS41FRUWzYsKHMxyQkJBAVFVVirH///sybN4/CwkJcXV1JSEhg4sSJpWreeOONMo9pNpv5/PPPOX/+PJGRkVfsNz8/v8Qu6Dk5OeW9PBGRWue/uzKIW7SN8wVmmtSvw7ujIggJ0l8qpeaqdGg6fPhwFbYBWVlZmM1m/P39S4z7+/uTkZFR5mMyMjLKrC8qKiIrK4vAwMAr1vz+mDt27CAyMpK8vDzq1avHkiVLCAkJuWK/06ZNY+rUqRV5iSIitYLFYvDmd/t587v9AES28mXmA2FW78skYq/s7pLRv5+9utLpv/Lqfz9uzTHbtWvHtm3b2LhxI48++iijR49m9+7dV3zeJ598kuzs7OLb0aNHy39hIiK1QG5eIY/MTywOTA/1aslHf+6uwCQO4Zo3t6wqfn5+ODs7l5oByszMLDVTdFlAQECZ9S4uLvj6+pZb8/tjurm5FS8Ej4iI4Oeff+bNN9/knXfeKfO53d3dcXd3t/4Fiog4uNSs84z5aAsHMs/h5uLEy0NDuTuima3bEqkydjPT5ObmRnh4OPHx8SXG4+Pjy7wYMEBkZGSp+lWrVhEREYGrq2u5NVc65mWGYZRYsyQiIle2Zm8md85Yz4HMc/h7u/PZI5EKTOJwqmSmaevWrYSGhuLmdm3Tr3FxcYwcOZKIiAgiIyOZO3cuaWlpxMTEAJdOiR0/fpyPPvoIuPRNuRkzZhAXF8eYMWNISEhg3rx5Jb4VN2HCBPr06cP06dMZMmQIS5cuZfXq1axfv7645qmnnmLgwIE0a9aM3NxcFi5cyJo1a1i5cuU1vR4REUdnGAZz1h7in//dg2FAWPP6zBkRTmNvXS9OHE+VhKYbb7yRlJQU2rZte03HiY6O5tSpU7zwwgukp6cTGhrKihUraNGiBQDp6emkpaUV1wcHB7NixQomTpzIzJkzCQoK4q233ireowmgZ8+eLFy4kGeeeYZnn32W1q1bs2jRouI9mgBOnjzJyJEjSU9Px8fHh86dO7Ny5Ur69et3Ta9HRMSRXSww88Ti7SxPPgHAfd2b8fydHXF30cXaxTFVep+m33JycmLPnj3XHJpqMu3TJCK1ydHTF3jk40R2p+fg4mRiyp0dGdGjareiEakO1bJPk4iI1E4JB08x9tOtnD5fgG9dN2Y9EEaPVtrNWxyfQpOIiFjFMAw+3HCYF79JwWwxCG3izdyREQTVr2Pr1kSqhUKTiIhcVV6hmWe/2snniZcu0j60axCvDu+Mh6vWL0ntodAkIiLlOpmTxyMfJ7Lt6FmcTPDkwA78pXew1i9JraPQJCIiV5R45Awx8xP5JTcfnzquzLi/G73bNLJ1WyI2odAkIiJlWrg5jWeX7qTQbNDO34u5o8Jp4VvX1m2J2EyVhKYpU6bg5+dXFYcSEREbKyiy8OLXu/l44xEABoYG8O+7u1DXXX/PltqtSvZpEu3TJCKOIetcPn+bv5XNh09jMsHf+7Vl7K03aP2SOCzt0yQiIhW241g2j3y8hRPZedRzd+GN6K7cEVL2BdNFaiOFJhER4auk40xavJ38Igut/Ooyd1QENzSuZ+u2ROyKQpOISC1WZLYwfeUe3v0xFYBb2zXijXu74VPH1cadidifKglNZ8+e5b///S/Hjx/HZDIRGBhI//79adCgQVUcXkREroOzFwoYtyCJH/dnATD21tbE9WuHs5PWL4mUxelaDzBv3jy6d+/Oxo0bsVgsmM1mNm7cyE033cS8efOqokcREaliezNyuXPGT/y4P4s6rs7MvD+Mf/Rvr8AkUo5r/vZcu3btSExMpF69kue+c3NzCQ8PZ9++fdfUYE2hb8+JSE2xcmc6cZ8lc6HATNMGdZg7MoKQIP3ektqpWr89ZzKZOHfuXKnQdO7cOX1FVUTEjlgsBm+s3sdb3x8AoGdrX2beH0aDum427kykZrjm0PTvf/+bvn37EhoaSpMmTQA4duwYu3bt4rXXXrvmBkVEpGLMFoPNqafJzM2jsZcH3YMbcqGgiImLtrE6JROAh3sF89Sg9rg4X/MqDZFao0o2tzSbzWzevJkTJ05gGAZNmjShe/fuODvXnqtf6/SciNiDlTvTmbp8N+nZecVjjeq54exkIiMnHzcXJ6YN68Tw8KY27FLEflz303OFhYUMGDCAunXrsnTpUpydnYmMjKxUsyIiUjVW7kzn0flb+f3fhH85VwBAfU9XPnyoO12a1a/23kQcQaXmZV1dXdm6dStpaWlatyQiYgfMFoOpy3eXCky/5ebsRGgTn2rrScTRVPpk9rBhwzhw4ADHjh2ryn5ERKQSNqeeLnFKriyZuflsTj1dTR2JOJ5KLwT38/OjqKiIsLAwhg8fTkBAQIlZp+eee65KGhQRkavLzC0/MFW0TkRKq/RCcCcnJ0wmE4ZhlHmKzmw2X3NzNYkWgouILSUcPMV97268at2CMTcR2dq3GjoSqRmqZZ+m5s2baz2TiIgdMFsMfjqQVW6NCQjwubT9gIhUTqVD0+HDh6uwDRERqYyzFwqIXbSNNXt/uWLN5b/eThkcosukiFyDKrlgr4iIVL/dJ3J4ZP4Wjp6+iIerE9Pu6kQdV+dS+zQF+HgwZXAIA0IDbditSM2n0CQiUgN9lXScyV9uJ6/QQtMGdXhnZDgdgy5tJ9AvJKDUjuCaYRK5dhUKTcHBwZVaxxQbG8v48eMr/DgRESmp0Gzh5W9S+GDDYQD6tG3EW/d2pb7n/64f5+xk0mJvkeugQqHpgw8+qNSTtGzZslKPExGR/8nMzeOxT5LYfPjSXkvjbruB2DvaahZJpJpUKDT17dv3evUhIiLlSDxyhr99ksjJnHzqubvw+j1diOoYYOu2RGqVCu0Ivn37diwWi9X1u3btoqioqMJNiYjIJYZh8PHGI9w7N4GTOfm0aVyPpY/1UmASsYEKhaZu3bpx6tQpq+sjIyNJS0urcFMiIgJ5hWb+8cV2nv1qJ4Vmg0GdAlgythetG9WzdWsitVKFTs8ZhsGzzz6Lp6enVfUFBQWVakpEpLY7duYCMfMT2Xk8BycTTBrQnr/2aaVNhUVsqEKhqU+fPuzdu9fq+sjISOrUqVPhpkREarMf9//C+AVJnLlQSANPV2bcH0avG/xs3ZZIrVeh0LRmzZrr1IaIiBiGwey1B/n3f/diMaBTEx9mjwijaQPrZvdF5PrS5pYiInbgXH4Rj3+WzMpdGQDcHd6UF4eG4uHqbOPOROSySocmi8XC+++/z3fffcfJkycxDKP4PpPJxHfffVclDYqIOLoDmed45OMtHPzlPK7OJp6/syP3d9dF0UXsTaVDU1xcHG+//TZAcWAymUwYhqE/6CIiVlq5M4PHP0/mXH4R/t7uzB4RTljzBrZuS0TKUOnQtGDBAgzDICgoiODgYFxcdKZPRMRaZovBa6v2MmvNQQC6Bzdk5v1hNPJyt3FnInIllU46ZrOZpk2bsn//ftzd9YdcRMRaZ84XMH5hEj/uzwLgzzcHM3lge1ydK7R1nohUs0r/Cb333nu5ePEihYWFVdmPiIhD23k8m8Ez1vPj/iw8XJ14896uPPvHEAUmkRqg0jNN9erVIycnh65du3LnnXdSv379Evc/99xz19qbiIhDWZx4jKeW7CC/yEILX0/mjAinQ6C3rdsSESuZjN9+7a0CnJycyl34bTabr7m5miQnJwcfHx+ys7Px9tYvQRH5n4IiCy99s5uPEo4AcGu7RrwR3Q0fT1cbdyYiFfn8rvRMU/Pm+jqsiMjVnMzJ42+fbCXxyBkAJtzehgm3t8HJSb8/RWqaSoemw4cPV2EbIiKO5+fDp/nbJ1v5JTcfLw8X3ojuyu0d/G3dlohUkvYJEBGpYoZh8OGGw7z0TQpFFoN2/l7MGRlOsF9dW7cmItfgmkJTQUEBP/30EydOnCi1hmnUqFHX1JiISE10scDM00t28GXScQD+2DmQ6cM7U9ddf0cVqekq/ad4//799OvXj6NHj5a6z2QyKTSJSK2TduoCj8xPJCU9B2cnE08ObM+fbw7W+k8RB1Hp0DR58mTS0tKqshcRkRprzd5MJizcRvbFQnzrujHj/jAiW/vaui0RqUKV3k1t/fr1uLi4EB8fD0C3bt1YsGABfn5+xWMiIo7OYjGY8f1+HvrgZ7IvFtKlWX2+Hn+zApOIA6p0aDpz5gwdOnTg9ttvx2Qy4erqSnR0NAEBAbzyyitV2aOIiF3KySvkkfmJ/HvVPgwD7uvenM8euYlAnzq2bk1EroNKn57z8vLCYrEAl3YH37NnD5s2bSItLY2DBw9WWYMiIvZo/8lcHvk4kUNZ53FzduKFIR25t3tzW7clItfRNW1uefDgQcxmM506dSIhIYGePXsCEBwcXGUNiojYktlisDn1NJm5eTT28qB7cEP+uyuDxz9P5kKBmUAfD2aPCKdrs/q2blVErrNKh6YHHniAtWvXsm/fPp5++mmGDRtGQUEBzs7OPP/881XYooiIbazcmc7U5btJz84rHqvr5sz5gktbrES28uXt+7vhV8/dVi2KSDWq9LXnfi81NZWkpCQ6duxIu3btquKQNYquPSfiWFbuTOfR+Vu50i/Ifh0aM3tEOC7OlV4aKiJ2oFquPQeQlZXF22+/zcaNG2nRogXjx49n06ZN1KlTh+bNdW5fRGoms8Vg6vLdVwxMADtP5Gj/JZFaptJ/RTp8+DBdunThpZdeYvXq1ezcuZPs7GwefPBBZsyYUemGZs2aRXBwMB4eHoSHh/Pjjz+WW7927VrCw8Px8PCgVatWzJkzp1TN4sWLCQkJwd3dnZCQEJYsWVLi/mnTpnHjjTfi5eVF48aNGTp0KHv37q30axCRmm1z6ukSp+TKkp6dx+bU09XUkYjYg0qHpieeeIL09HSaNGnC5TN8vXr1wtvbu9L7NC1atIjY2FiefvppkpKS6N27NwMHDrziJpqpqakMGjSI3r17k5SUxFNPPcX48eNZvHhxcU1CQgLR0dGMHDmS5ORkRo4cyT333MOmTZuKa9auXcvYsWPZuHEj8fHxFBUVERUVxfnz5yv1OkSkZsvMLT8wVbRORBxDpdc0NWzYEBcXF1JTU/Hy8uKmm25iw4YNdOnShbS0NM6cOVPhY/bo0YOwsDBmz55dPNahQweGDh3KtGnTStVPmjSJZcuWkZKSUjwWExNDcnIyCQkJAERHR5OTk8O3335bXDNgwAAaNGjAggULyuzjl19+oXHjxqxdu5Y+ffpY1bvWNIk4jvhdJxnz8Zar1i0Yc5M2sRSp4Sry+V3pmaaLFy8SEBBA3bolr9p97tw58vPzK3y8goICEhMTiYqKKjEeFRXFhg0bynxMQkJCqfr+/fuzZcsWCgsLy6250jEBsrOzgUvB8Ery8/PJyckpcRORmm9PRg4vfrO73BoTEOhzafsBEak9Kh2aWrduza5du5g/fz5wKUS8/fbbpKam0rZt2wofLysrC7PZjL+/f4lxf39/MjIyynxMRkZGmfVFRUVkZWWVW3OlYxqGQVxcHDfffDOhoaFX7HfatGn4+PgU35o1a3bV1ygi9m3lznTumrWBtNMX8KvnBlwKSL91+ecpg0NwdtJCcJHapNKhacyYMRiGwejRozGZTGzbto3Y2FhMJhMPP/xwpRv6/bdRDMMo9xsqZdX/frwix3zsscfYvn37FU/dXfbkk0+SnZ1dfDt69Gi59SJivywWg9dX7SVm/lYuFJjp2dqX+Il9mTMijAAfjxK1AT4ezB4RxoDQQBt1KyK2UuktB8aPH8/evXt55513SgSVMWPGMH78+Aofz8/PD2dn51IzQJmZmaVmii4LCAgos97FxQVfX99ya8o65rhx41i2bBnr1q2jadOm5fbr7u6Ou7s2tBOp6XLzCpm4KJnVKScBeLhXME8Nao+LsxMDQgPpFxJQakdwzTCJ1E6VDk0mk4lZs2bxxBNPsGXLFgzDICIiotKXUHFzcyM8PJz4+HiGDRtWPB4fH8+QIUPKfExkZCTLly8vMbZq1SoiIiJwdXUtromPj2fixIklai5f8gUuzTyNGzeOJUuWsGbNGl0GRqSWOPTLOf76cSIHMs/h5uLEtGGdGB5e8i9Mzk4mLfYWEeAaN7c0m80UFBTg7++PYRgcPXq0+DSVtd86+624uDhGjhxJREQEkZGRzJ07l7S0NGJiYoBLp8SOHz/ORx99BFz6ptyMGTOIi4tjzJgxJCQkMG/evBKn1iZMmECfPn2YPn06Q4YMYenSpaxevZr169cX14wdO5ZPP/2UpUuX4uXlVTwz5ePjQ506ulq5iCP6YW8m4xckkZtXRIC3B3NG6vpxInIVRiX99NNPRosWLQwnJ6dSN2dn58oe1pg5c6bRokULw83NzQgLCzPWrl1bfN/o0aONvn37lqhfs2aN0a1bN8PNzc1o2bKlMXv27FLH/Pzzz4127doZrq6uRvv27Y3FixeXuB8o8/b+++9b3Xd2drYBGNnZ2RV6vSJSvSwWizHrhwNGy8lfGy0mfW3cNesn42TORVu3JSI2UpHP70rv09S1a1e2b99e5n0mkwmz2VypEFdTaZ8mEft3scDME4u3szz5BAD3dW/G83d2xN3F2cadiYitVMu15w4cOECDBg34/PPPadWqla7BJCJ27diZC/z1o0R2p+fg4mRiyp0dGdGjuX53iYjVKh2a+vbty44dO+jTpw8uLte0NEpE5LpKOHiKsZ9u5fT5AnzrujHrgTB6tNLibhGpmEqfnsvIyOCWW27Bzc2NqKioUlNazz33XJU0WFPo9JyI/TEMg48SjvDC17sxWwxCm3jzzsgImtTXFzxE5JJqOT23cuVKDh48iMViYdeuXaXur22hSUTsS36RmWe/2slnW44BMKRrEK/e1Zk6blq/JCKVU+nQ9Mwzz9S6xd4iUjNk5uTxyPxEktLO4mSCyQPbM6a31l6KyLWp9GVUzp07R2BgIPv376ewsBCLxVLiJiJiC0lpZ/jj2+tJSjuLt4cL7z/Unb/2aa3AJCLXrNKh6c9//jNFRUU0btwYZ2dNd4uI7X225SjR72wkMzefNo3rseyxm+nbtpGt2xIRB1Hp03O//PILOTk5tGnThptvvrnE4imTycS8efOqpEERkaspNFt4+ZsUPthwGICoEH9ej+5KPXd9s1dEqk6lvz3n5OSEyWTCMIwS096Xf65t65307TkR2zh9voCxn2wl4dApAGLvaMP429rgpIvqiogVquXbc82ba1M4EbGt3Sdy+OvHWzh25iJ13Zx5Pbor/TsG2LotEXFQlQ5Nhw8frsI2REQqZnnyCf7xRTJ5hRZa+Hry7qgI2vp72botEXFgOuEvIjWK2WLw2qq9zFpzEIDebfyYcV8YPp6uNu5MRBydQpOI1BjZFwuJXZjED3t/AeCRPq14YkB7nLV+SUSqgUKTiNQIBzLP8dePtnAo6zzuLk5MH96Zod2a2LotEalFFJpExO59l3KS2IXbyM0vIsjHg3dGRtCpqY+t2xKRWkahSUTslmEYzPzhAK/F78MwoHvLhswaEYZfPXdbtyYitZBCk4jYpfP5Rfzji2RW7MgAYMRNzXnujx1xc6n0hQxERK6JQpOI2J2jpy8w5qMt7MnIxdXZxAtDQrmve3NbtyUitZxCk4jYlZ8OZDH2062cvVCIXz135owII6JlQ1u3JSKi0CQi9sEwDN776TCvrEjBbDHo0tSHOSPDCfSpY+vWREQAhSYRsQN5hWaeXrKTxVuPAXBXWBNeGdYJD1dnG3cmIvI/Ck0iUm3MFoPNqafJzM2jsZcH3YMb8ktuPo98vIXkY9k4O5l4alAHHu7VUte2FBG7o9AkItVi5c50pi7fTXp2XvGYb103Cs0WcvKKqO/pysz7w+h1g58NuxQRuTKFJhG57lbuTOfR+Vsxfjd+6nwBAE3qe7BgTCTNfT2rvzkREStpwxMRua7MFoOpy3eXCky/VWQxaNJAC75FxL4pNInIdbU59XSJU3JlOZmTz+bU09XUkYhI5Sg0ich1lZlbfmCqaJ2IiK0oNInIddXYy6NK60REbEULwUXkuikyW1idcrLcGhMQ4HNp+wEREXum0CQi18WZ8wWMW5DE+gNZV6y5vBPTlMEhODtpXyYRsW8KTSJS5VLSc/jrx1s4evoinm7O/PvuLjiZKLVPU4CPB1MGhzAgNNCG3YqIWEehSUSq1Dfb03n882QuFppp3tCTuaPCaR/gDUC/kIBSO4JrhklEagqFJhGpEmaLwevxe5n5w0EAerfx4+37ulHf0624xtnJRGRrX1u1KCJyTRSaROSaZV8sJHZhEj/s/QWAv/ZpxRP92+HirC/oiojjUGgSkWtyIDOXMR8lkpp1HncXJ6YP78zQbk1s3ZaISJVTaBKRSovffZKJi7ZxLr+IIB8P5o6KILSJj63bEhG5LhSaRKTCLBaDt78/wP+t3gdAj+CGzHwgDL967jbuTETk+lFoEpEKOZdfxN8/28Z/d13atHJ0ZAue+WMIrlq/JCIOTqFJRKx2OOs8Yz7awv7Mc7g5O/HS0FDuubGZrdsSEakWCk0iYpU1ezMZvyCJnLwi/L3dmTMinG7NG9i6LRGRaqPQJCLlMgyDd9Yd4p8r92AxIKx5feaMCKexty6wKyK1i0KTiFzRhYIinvhiO19vTwfg3hubMXVIR9xdnG3cmYhI9VNoEpEyHT19gb9+nEhKeg4uTiaev7MjD/Rojsmky56ISO2k0CQipWw4kMXYT7dy5kIhfvXcmPVAON2DG9q6LRERm1JoEpFihmHw/k+HeXlFCmaLQacmPrwzMpyg+nVs3ZqIiM0pNIkIAHmFZp5espPFW48BcFe3JrxyVyc8XLV+SUQEFJpEBEjPvkjMx4kkH8vG2cnEU4M68HCvllq/JCLyGwpNIrXcz4dP8+j8RLLOFVDf05WZ94fR6wY/W7clImJ3FJpEarFPNh3h+WW7KDQbtA/w4t1RETRr6GnrtkRE7JJCk0gtVFBkYcqyXSzYnAbAHzoH8q8/dcbTTb8SRESuRL8hRWqZzJw8Hv1kK4lHzmAywRP92xPTt5XWL4mIXIVCk0gtsu3oWR75eAsnc/Lx8nDhrfu6cWu7xrZuS0SkRlBoEqklPt9ylKe/2klBkYUbGtfj3VERBPvVtXVbIiI1hkKTiIMrNFt4+ZsUPthwGIB+If68fk8XvDxcbduYiEgNo9Ak4sBOncvnsU+TSDh0CoDYO9ow/rY2ODlp/ZKISEU52bqB35s1axbBwcF4eHgQHh7Ojz/+WG792rVrCQ8Px8PDg1atWjFnzpxSNYsXLyYkJAR3d3dCQkJYsmRJifvXrVvH4MGDCQoKwmQy8dVXX1XlSxKxiZ3Hs7lzxk8kHDpFXTdn5o4MJ/aOtgpMIiKVZFehadGiRcTGxvL000+TlJRE7969GThwIGlpaWXWp6amMmjQIHr37k1SUhJPPfUU48ePZ/HixcU1CQkJREdHM3LkSJKTkxk5ciT33HMPmzZtKq45f/48Xbp0YcaMGdf9NYpUh6XbjvOnORs4fvYiLX09+WpsL6I6Bti6LRGRGs1kGIZh6yYu69GjB2FhYcyePbt4rEOHDgwdOpRp06aVqp80aRLLli0jJSWleCwmJobk5GQSEhIAiI6OJicnh2+//ba4ZsCAATRo0IAFCxaUOqbJZGLJkiUMHTq0Qr3n5OTg4+NDdnY23t7eFXqsSGWYLQabU0+TmZtHYy8Pugc3BOCf/93DO2sPAXBLu0a8eW83fOpo/ZKISFkq8vltN2uaCgoKSExMZPLkySXGo6Ki2LBhQ5mPSUhIICoqqsRY//79mTdvHoWFhbi6upKQkMDEiRNL1bzxxhtV2r9IdVq5M52py3eTnp1XPObv5Y5vPTd2p+cC8OgtrXk8qh3OOh0nIlIl7CY0ZWVlYTab8ff3LzHu7+9PRkZGmY/JyMgos76oqIisrCwCAwOvWHOlY1orPz+f/Pz84p9zcnKu6Xgi1lq5M51H52/l91PEJ3PzOZmbj5uzE69Hd+GPnYNs0p+IiKOyqzVNQKldiQ3DKHen4rLqfz9e0WNaY9q0afj4+BTfmjVrdk3HE7GG2WIwdfnuUoHpt7zruDAwNLDaehIRqS3sJjT5+fnh7OxcagYoMzOz1EzRZQEBAWXWu7i44OvrW27NlY5prSeffJLs7Ozi29GjR6/peCLW2Jx6usQpubJknStgc+rpaupIRKT2sJvQ5ObmRnh4OPHx8SXG4+Pj6dmzZ5mPiYyMLFW/atUqIiIicHV1LbfmSse0lru7O97e3iVuItdbZm75gamidSIiYj27WdMEEBcXx8iRI4mIiCAyMpK5c+eSlpZGTEwMcGl25/jx43z00UfApW/KzZgxg7i4OMaMGUNCQgLz5s0r8a24CRMm0KdPH6ZPn86QIUNYunQpq1evZv369cU1586d48CBA8U/p6amsm3bNho2bEjz5s2r6dWLXF1jL48qrRMREevZVWiKjo7m1KlTvPDCC6SnpxMaGsqKFSto0aIFAOnp6SX2bAoODmbFihVMnDiRmTNnEhQUxFtvvcXw4cOLa3r27MnChQt55plnePbZZ2ndujWLFi2iR48exTVbtmzh1ltvLf45Li4OgNGjR/PBBx9c51ctYp3z+UUs3nqs3BoTEODzv+0HRESk6tjVPk01mfZpkusp+ehZJixM4vCpC5hMUNaf2stfbZg9IowBWgguImKVGrlPk4iUZrEYzP3xEP/+716KLAZBPh78X3RXzlwoKLVPU4CPB1MGhygwiYhcJwpNInYqIzuPuM+2seHgpYvtDuoUwLRhnfHxvPQlh34hAaV2BNdGliIi149Ck4gdWrUrg0mLt3PmQiF1XJ2ZemdH7o5oWmJ/MWcnE5GtfW3YpYhI7aLQJGJHLhaYeemb3Xyy6dIXHkKbePPmvd1o3aiejTsTERGFJhE7sftEDuMXJnEg8xwAj/Rpxd+j2uHmYjfbqYmI1GoKTSI2ZhgG7/90mFe/3UOB2UJjL3dev6crN7fxs3VrIiLyGwpNIjaUdS6fxz9PZs3eXwC4o0Njpg/vjG89dxt3JiIiv6fQJGIja/Zm8vjnyWSdK8DdxYln/tCBETe1uOaLSYuIyPWh0CRSzfKLzEz/di/v/ZQKQDt/L966rxvtArxs3JmIiJRHoUmkGh3IzGXcgm2kpOcA8GDPlkwe2B4PV2cbdyYiIlej0CRSDQzD4NPNabz49W7yCi00rOvGv+/uzG3t/W3dmoiIWEmhSeQ6O3O+gMlfbue/u04C0LuNH6/d3YXG3h427kxERCpCoUnkOtpwIIu4z5LJyMnD1dnEpAHtebhXME663ImISI2j0CRyHRSaLbwev485aw9iGNCqUV3eurcboU18bN2aiIhUkkKTSBU7nHWeCQuTSD6WDcB93Zvx7B9D8HTTHzcRkZpMv8VFqohhGCzeepwpS3dyvsCMTx1XXr2rEwM7Bdq6NRERqQIKTSJVIPtiIc98tZPlyScA6BHckP+L7kpQ/To27kxERKqKQpPINdpy+DQTFm7j+NmLODuZiOvXlpi+rXHWYm8REYei0CRSSUVmCzN+OMBb3+3HYkDzhp68eW9XujVvYOvWRETkOlBoEqmEY2cuELtwG1uOnAHgrm5NmDqkI14erjbuTERErheFJpEKWp58gqeW7CA3r4h67i68NDSUod2a2LotERG5zhSaRKx0Lr+I55ft4ovEYwB0a16fN6O70dzX08adiYhIdVBoErFC8tGzTFiYxOFTF3AywWO33sD429vg4uxk69ZERKSaKDSJlMNiMXhn3SFeW7WXIotBkI8H/xfdlR6tfG3dmoiIVDOFJhHAbDHYnHqazNw8Gnt50D24Ib/k5hP32TY2HDwFwKBOAUwb1hkfTy32FhGpjRSapNZbuTOdqct3k56dVzzWwNOVgiIL5wvM1HF1ZuqdHbk7oikmk/ZeEhGprRSapFZbuTOdR+dvxfjd+JkLhQA0b1iH9x/qTutG9aq/ORERsStaxSq1ltliMHX57lKB6bcKiiy09K1bbT2JiIj9UmiSWmtz6ukSp+TKkpGTz+bU09XUkYiI2DOFJqm19mTkWFWXmVt+sBIRkdpBa5qk1jlx9iIzfzjAwp+PWlXf2MvjOnckIiI1gUKT1BoZ2XnMWnOAhZuPUmC2AODm7FT8779nAgJ8Lm0/ICIiotAkDi8zJ49Zaw7y6eY0CoouBaTuwQ2J69eWsxcKeHT+VoASC8IvbywwZXAIzk7aZkBERBSaxIH9kpvPnLUHmb/xCPm/hqUbWzZg4h1tiWztW7zn0uwRYaX2aQrw8WDK4BAGhAbapHcREbE/Ck3icLLO5TN33SE+SjhMXuGlsBTWvD5x/drR6wbfUhtUDggNpF9IQKkdwTXDJCIiv6XQJA7j9PkC5q47xIcbDnOx0AxAl2b1ievXlj5t/MrdzdvZyURka11PTkRErkyhSWq8sxcKePfHQ3zw02HOF1wKS52a+BDXry23tGukS5+IiEiVUGiSGiv7QiHz1h/ivZ8Ocy6/CICOQd5MvKMtt3dorLAkIiJVSqFJapycvELeW5/KvPWp5OZdCksdAr2JvaMNUSH+CksiInJdKDRJjZGbV8j7Px3mPz8eIufXsNTO34vYO9rQv2MATlq4LSIi15FCk9i9c/lFfLjhMO/+eIizFwoBaNO4HhPuaMOg0ECFJRERqRYKTWK3zucX8VHCEeauO8iZX8NSq0Z1ib2jLX/oFKgtAUREpFopNInduVhg5uONh3ln7SFOnS8AINivLhNub8PgLkEKSyIiYhMKTWI38grNzN94hDlrD5F1Lh+AFr6ejL+tDUO6BuHi7GTjDkVEpDZTaBKbyys0s2BzGrPXHCQz91JYatawDuNua8Owbk1wVVgSERE7oNAkNpNfZGbRz0eZ9cNBMnIuXfetSf06jLvtBoaHN1VYEhERu6LQJNWuoMjCZ1uOMvOHA8UXyQ3y8WDsbTdwd3gz3FwUlkRExP4oNEmVM1uMMi9+W2i28EXiMWZ8f4DjZy8CEODtwdhbW3PPjc1wd3G2ceciIiJXptAkVWrlznSmLt9dPIMEEODtTr8Qf9bs+4Wjpy+FpcZe7vztltbc2705Hq4KSyIiYv8UmqTKrNyZzqPzt2L8bjwjJ5+PN6YB4FfPnUdvac0DPRSWRESkZlFokiphthhMXb67VGD6LS8PF9Y8fgv1PPS/nYiI1Dz69JJKy75QSEpGDinpOazd+0uJU3Jlyc0rYsfxbCJb+1ZThyIiIlVHoUmuymIxOHzqPCnpuaSk57AnI4eU9NzixdwVkZlbfrASERGxVwpNUkJuXiF7My6Fo92/hqS9GblcLDSXWd+kfh06BHrj7eHCl0nHr3r8xl4eVd2yiIhItVBoqqUsFoNjZy6yO/3S6bWU9BxSMnKKv932ex6uTrTz96JDoHfxrX2gF94ersClNU0Jh06RkZ1X5romExDgc2n7ARERkZpIocnOXWnPo4q4UFDEnoxc9vw6c3TpFFsu5/KLyqwP9PH4NRh50T7gUkAK9qtb7vM6O5mYMjiER+dvxQQlgtPlR00ZHKKL7YqISI1ld6Fp1qxZ/Otf/yI9PZ2OHTvyxhtv0Lt37yvWr127lri4OHbt2kVQUBBPPPEEMTExJWoWL17Ms88+y8GDB2ndujUvv/wyw4YNu6bnrQ5l7XkU6OPBlMEhDAgNLFVvGAYnsvNIOfG/maOU9FwOnzqPUcb0j5uzE2386/1m9siLDgHeNKjrVql+B4QGMntEWOl9msrpWUREpKawq9C0aNEiYmNjmTVrFr169eKdd95h4MCB7N69m+bNm5eqT01NZdCgQYwZM4b58+fz008/8be//Y1GjRoxfPhwABISEoiOjubFF19k2LBhLFmyhHvuuYf169fTo0ePSj1vdbjinkfZeTw6fytv3tuVln51f505+t8MUk5e2bNHjbzci4NRSOD/Zo+q+vpuA0ID6RcScM2zYyIiIvbGZBhlzUHYRo8ePQgLC2P27NnFYx06dGDo0KFMmzatVP2kSZNYtmwZKSkpxWMxMTEkJyeTkJAAQHR0NDk5OXz77bfFNQMGDKBBgwYsWLCgUs9blpycHHx8fMjOzsbb27tiL/x3zBaDm6d/f9Wv8JfFxcnEDY3rFQej9oGX1iH51XO/pp5EREQcUUU+v+1mpqmgoIDExEQmT55cYjwqKooNGzaU+ZiEhASioqJKjPXv35958+ZRWFiIq6srCQkJTJw4sVTNG2+8UennBcjPzyc/P7/455ycnKu+RmttTj1tVWDy8nChc1MfOgT8b3H2DY3r6YK3IiIi14HdhKasrCzMZjP+/v4lxv39/cnIyCjzMRkZGWXWFxUVkZWVRWBg4BVrLh+zMs8LMG3aNKZOnWr166sIa/cyenFIR4Z2a3pdehAREZGS7G5KwmQqufbFMIxSY1er//24Nces6PM++eSTZGdnF9+OHj16xdqKsnYvI3/vOlX2nCIiIlI+u5lp8vPzw9nZudTsTmZmZqlZoMsCAgLKrHdxccHX17fcmsvHrMzzAri7u+Pufn3WCXUPbkigj4f2PBIREbEjdjPT5ObmRnh4OPHx8SXG4+Pj6dmzZ5mPiYyMLFW/atUqIiIicHV1Lbfm8jEr87zX2+U9j+B/exxdpj2PREREbMNuQhNAXFwc//nPf3jvvfdISUlh4sSJpKWlFe+79OSTTzJq1Kji+piYGI4cOUJcXBwpKSm89957zJs3j8cff7y4ZsKECaxatYrp06ezZ88epk+fzurVq4mNjbX6eW3h8p5HAT4lT9UF+Hgwe0SY9jwSERGpboadmTlzptGiRQvDzc3NCAsLM9auXVt83+jRo42+ffuWqF+zZo3RrVs3w83NzWjZsqUxe/bsUsf8/PPPjXbt2hmurq5G+/btjcWLF1foea2RnZ1tAEZ2dnaFHnc1RWaLseFAlvFV0jFjw4Eso8hsqdLji4iI1GYV+fy2q32aarKq3KdJREREqkdFPr/t6vSciIiIiL1SaBIRERGxgkKTiIiIiBUUmkRERESsoNAkIiIiYgWFJhERERErKDSJiIiIWEGhSURERMQKCk0iIiIiVnCxdQOO4vLG6jk5OTbuRERERKx1+XPbmgukKDRVkdzcXACaNWtm405ERESkonJzc/Hx8Sm3RteeqyIWi4UTJ07g5eWFyWSq0mPn5OTQrFkzjh49quvaXUd6n6uH3ufqofe5euh9rj7X6702DIPc3FyCgoJwcip/1ZJmmqqIk5MTTZs2va7P4e3trT+U1UDvc/XQ+1w99D5XD73P1ed6vNdXm2G6TAvBRURERKyg0CQiIiJiBYWmGsDd3Z0pU6bg7u5u61Ycmt7n6qH3uXrofa4eep+rjz2811oILiIiImIFzTSJiIiIWEGhSURERMQKCk0iIiIiVlBoEhEREbGCQpOdmzVrFsHBwXh4eBAeHs6PP/5o65YcyrRp07jxxhvx8vKicePGDB06lL1799q6LYc3bdo0TCYTsbGxtm7FIR0/fpwRI0bg6+uLp6cnXbt2JTEx0dZtOZSioiKeeeYZgoODqVOnDq1ateKFF17AYrHYurUabd26dQwePJigoCBMJhNfffVVifsNw+D5558nKCiIOnXqcMstt7Br165q60+hyY4tWrSI2NhYnn76aZKSkujduzcDBw4kLS3N1q05jLVr1zJ27Fg2btxIfHw8RUVFREVFcf78eVu35rB+/vln5s6dS+fOnW3dikM6c+YMvXr1wtXVlW+//Zbdu3fz2muvUb9+fVu35lCmT5/OnDlzmDFjBikpKfzzn//kX//6F2+//batW6vRzp8/T5cuXZgxY0aZ9//zn//k9ddfZ8aMGfz8888EBATQr1+/4uu/XneG2K3u3bsbMTExJcbat29vTJ482UYdOb7MzEwDMNauXWvrVhxSbm6u0aZNGyM+Pt7o27evMWHCBFu35HAmTZpk3HzzzbZuw+H94Q9/MB5++OESY3fddZcxYsQIG3XkeABjyZIlxT9bLBYjICDAePXVV4vH8vLyDB8fH2POnDnV0pNmmuxUQUEBiYmJREVFlRiPiopiw4YNNurK8WVnZwPQsGFDG3fimMaOHcsf/vAH7rjjDlu34rCWLVtGREQEd999N40bN6Zbt268++67tm7L4dx8881899137Nu3D4Dk5GTWr1/PoEGDbNyZ40pNTSUjI6PE56K7uzt9+/atts9FXbDXTmVlZWE2m/H39y8x7u/vT0ZGho26cmyGYRAXF8fNN99MaGiordtxOAsXLmTr1q38/PPPtm7FoR06dIjZs2cTFxfHU089xebNmxk/fjzu7u6MGjXK1u05jEmTJpGdnU379u1xdnbGbDbz8ssvc99999m6NYd1+bOvrM/FI0eOVEsPCk12zmQylfjZMIxSY1I1HnvsMbZv38769ett3YrDOXr0KBMmTGDVqlV4eHjYuh2HZrFYiIiI4JVXXgGgW7du7Nq1i9mzZys0VaFFixYxf/58Pv30Uzp27Mi2bduIjY0lKCiI0aNH27o9h2bLz0WFJjvl5+eHs7NzqVmlzMzMUilbrt24ceNYtmwZ69ato2nTprZux+EkJiaSmZlJeHh48ZjZbGbdunXMmDGD/Px8nJ2dbdih4wgMDCQkJKTEWIcOHVi8eLGNOnJM//jHP5g8eTL33nsvAJ06deLIkSNMmzZNoek6CQgIAC7NOAUGBhaPV+fnotY02Sk3NzfCw8OJj48vMR4fH0/Pnj1t1JXjMQyDxx57jC+//JLvv/+e4OBgW7fkkG6//XZ27NjBtm3bim8RERE88MADbNu2TYGpCvXq1avUthn79u2jRYsWNurIMV24cAEnp5Ifoc7Oztpy4DoKDg4mICCgxOdiQUEBa9eurbbPRc002bG4uDhGjhxJREQEkZGRzJ07l7S0NGJiYmzdmsMYO3Ysn376KUuXLsXLy6t4Zs/Hx4c6derYuDvH4eXlVWqdWN26dfH19dX6sSo2ceJEevbsySuvvMI999zD5s2bmTt3LnPnzrV1aw5l8ODBvPzyyzRv3pyOHTuSlJTE66+/zsMPP2zr1mq0c+fOceDAgeKfU1NT2bZtGw0bNqR58+bExsbyyiuv0KZNG9q0acMrr7yCp6cn999/f/U0WC3f0ZNKmzlzptGiRQvDzc3NCAsL01fhqxhQ5u3999+3dWsOT1sOXD/Lly83QkNDDXd3d6N9+/bG3Llzbd2Sw8nJyTEmTJhgNG/e3PDw8DBatWplPP3000Z+fr6tW6vRfvjhhzJ/J48ePdowjEvbDkyZMsUICAgw3N3djT59+hg7duyotv5MhmEY1RPPRERERGourWkSERERsYJCk4iIiIgVFJpERERErKDQJCIiImIFhSYRERERKyg0iYiIiFhBoUlERETECgpNIlLjrVmzBpPJxNmzZ23diog4MIUmEalxbrnlFmJjY4t/7tmzJ+np6fj4+NisJwU3Ecena8+JSI3n5uZWfAV0EZHrRTNNIlKjPPjgg6xdu5Y333wTk8mEyWTigw8+KDHL88EHH1C/fn2+/vpr2rVrh6enJ3/60584f/48H374IS1btqRBgwaMGzcOs9lcfOyCggKeeOIJmjRpQt26denRowdr1qwpvv/IkSMMHjyYBg0aULduXTp27MiKFSs4fPgwt956KwANGjTAZDLx4IMPAmAYBv/85z9p1aoVderUoUuXLnzxxRfFx7w8Q/XNN9/QpUsXPDw86NGjBzt27Lju76WIVIxmmkSkRnnzzTfZt28foaGhvPDCCwDs2rWrVN2FCxd46623WLhwIbm5udx1113cdddd1K9fnxUrVnDo0CGGDx/OzTffTHR0NAAPPfQQhw8fZuHChQQFBbFkyRIGDBjAjh07aNOmDWPHjqWgoIB169ZRt25ddu/eTb169WjWrBmLFy9m+PDh7N27F29vb+rUqQPAM888w5dffsns2bNp06YN69atY8SIETRq1Ii+ffsW9/uPf/yDN998k4CAAJ566inuvPNO9u3bh6urazW8qyJilWq7NLCISBXp27evMWHChOKfL18Z/cyZM4ZhGMb7779vAMaBAweKax555BHD09PTyM3NLR7r37+/8cgjjxiGYRgHDhwwTCaTcfz48RLPdfvttxtPPvmkYRiG0alTJ+P5558vs6ff92AYhnHu3DnDw8PD2LBhQ4naP//5z8Z9991X4nELFy4svv/UqVNGnTp1jEWLFln5johIddBMk4g4JE9PT1q3bl38s7+/Py1btqRevXolxjIzMwHYunUrhmHQtm3bEsfJz8/H19cXgPHjx/Poo4+yatUq7rjjDoYPH07nzp2v2MPu3bvJy8ujX79+JcYLCgro1q1bibHIyMjif2/YsCHt2rUjJSWlgq9aRK4nhSYRcUi/P61lMpnKHLNYLABYLBacnZ1JTEzE2dm5RN3loPWXv/yF/v37880337Bq1SqmTZvGa6+9xrhx48rs4fKxv/nmG5o0aVLiPnd396u+BpPJdNUaEak+Ck0iUuO4ubmVWMBdFbp164bZbCYzM5PevXtfsa5Zs2bExMQQExPDk08+ybvvvsu4ceNwc3MDKNFXSEgI7u7upKWllVi/VJaNGzfSvHlzAM6cOcO+ffto3759FbwyEakqCk0iUuO0bNmSTZs2cfjwYerVq1c8o3Mt2rZtywMPPMCoUaN47bXX6NatG1lZWXz//fd06tSJQYMGERsby8CBA2nbti1nzpzh+++/p0OHDgC0aNECk8nE119/zaBBg6hTpw5eXl48/vjjTJw4EYvFws0330xOTg4bNmygXr16jB49uvj5X3jhBXx9ffH39+fpp5/Gz8+PoUOHXvPrEpGqoy0HRKTGefzxx3F2diYkJIRGjRqRlpZWJcd9//33GTVqFH//+99p164dd955J5s2baJZs2bApVmksWPH0qFDBwYMGEC7du2YNWsWAE2aNGHq1KlMnjwZf39/HnvsMQBefPFFnnvuOaZNm0aHDh3o378/y5cvJzg4uMRzv/rqq0yYMIHw8HDS09NZtmxZ8eyViNgHk2EYhq2bEBGprdasWcOtt97KmTNnqF+/vq3bEZFyaKZJRERExAoKTSIiIiJW0Ok5EREREStopklERETECgpNIiIiIlZQaBIRERGxgkKTiIiIiBUUmkRERESsoNAkIiIiYgWFJhERERErKDSJiIiIWEGhSURERMQK/w+LC3fpotQinwAAAABJRU5ErkJggg==",
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -880,7 +1068,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHFCAYAAADv8c1wAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABK6klEQVR4nO3deXxU1f3/8ffNHrIRtmxACApIDAJJBMPqGpZqRdEiFQS1tKgVA1oVN3D5FvWnbV0gKsWFaoFWXLCiEiugSJAdEVBQgmFJjAHMhC3L5P7+oBkdMkkmySSz5PV8PObxYM6cufdcriYfzj3n8zFM0zQFAACAOvm5ewAAAADegKAJAADACQRNAAAATiBoAgAAcAJBEwAAgBMImgAAAJxA0AQAAOAEgiYAAAAnEDQBAAA4gaAJgFcyDMOp16pVq5p0ntmzZ8swjHr7HThwQFlZWRo+fLjatm0rwzD06quvOux7//33q3///mrXrp1CQkLUvXt3/f73v9f333/fpLECaF4B7h4AADRGbm6u3ftHH31UK1eu1CeffGLXnpyc3KTz/O53v9PIkSPr7fftt9/qjTfeUL9+/TR69GgtWrSo1r4//fSTxo8fr969eysiIkI7d+7UY489pmXLlmnHjh1q3759k8YMoHkY1J4D4AsmT56sN998U8eOHauz34kTJ9SmTRuXn7+qqkp+fqcn7zdu3Kjzzz9fr7zyiiZPnuzU9z/44AONHj1aCxYs0E033eTy8QFoOh7PAfBZF154oVJSUvTpp59q0KBBatOmjS0gWbJkiTIzMxUXF6fQ0FD17t1b9957r44fP253DGcfz1UHTI3VsWNHSVJAAA8AAE/F/50AfFpBQYEmTJigu+++W3/+859twc2ePXs0evRoZWVlKSwsTF9//bWeeOIJrV+/vsYjvuZSWVmpiooKff3118rKylLPnj119dVXt8i5ATQcQRMAn3bkyBH9+9//1sUXX2zX/sADD9j+bJqmBg8erN69e2v48OH68ssvdd555zXruAoLCxUXF2d7P3DgQK1cuVLh4eHNel4AjcfjOQA+LTo6ukbAJEl79+7Vb3/7W8XGxsrf31+BgYEaPny4JGnXrl3NPq4OHTpow4YNWrNmjebPn68jR47ooosuUkFBQbOfG0DjMNMEwKf9cjan2rFjxzR06FCFhIToscceU8+ePdWmTRvt379fV199tU6ePNns4woICFB6erokafDgwRo5cqSSkpL0+OOP65lnnmn28wNoOIImAD7N0SLuTz75RIcOHdKqVatss0vS6VQA7tK5c2fFx8dr9+7dbhsDgLrxeA5Aq1MdSAUHB9u1v/jii+4YjqTTeZ4OHDigs88+221jAFA3ZpoAtDqDBg1SdHS0pk6dqlmzZikwMFBvvPGGtm3b1qTjvvnmm5JOr5eSTudrql7Yfc0110iSvvzyS02fPl3XXHONunfvLj8/P23fvl1//etf1b59e911111NGgOA5kPQBKDVad++vd5//33deeedmjBhgsLCwnTllVdqyZIlSk1NbfRxr732Wrv3c+fO1dy5cyWd3qEnSTExMYqPj9fTTz+tgoICVVZWqnPnzrr88st13333qUuXLo2/MADNiozgAFCH6dOn6x//+IeKi4vdPRQAbsZMEwA4UFRUpNzcXL311lvKyMhw93AAeAAWggOAA8uXL9f111+vHj16kAIAgCQezwEAADiFmSYAAAAnEDQBAAA4gaAJAADACeyec5GqqiodOnRIERERDss2AAAAz2OapkpLSxUfHy8/v7rnkgiaXOTQoUMkpQMAwEvt379fnTt3rrMPQZOLRERESDr9lx4ZGenm0QAAAGdYLBZ16dLF9nu8LgRNLlL9SC4yMpKgCQAAL+PM0hoWggMAADiBoAkAAMAJBE0AAABOIGgCAABwAkETAACAEwiaAAAAnEDQBAAA4ASCJgAAACcQNAEAADiBjOAAAMCjWatMrc87oqLSU+oUEaIBSe3k71d/Bm9XI2gCAAAe68OvCvTweztVUHLK1hYXFaJZVyRrZEpci46Fx3MAAMAjffhVgW55fbNdwCRJhSWndMvrm/XhVwUtOh6CJgAA4HGsVaYefm+nTAefVbc9/N5OWasc9WgeBE0AAMDjrM87UmOG6ZdMSQUlp7Q+70iLjYmgCQAAeJyi0toDpsb0cwWCJgAA4FHKK6u08usip/p2ighp5tH8jN1zAADAY3xTWKrpS7ZqZ4Glzn6GpNio0+kHWgpBEwAAaFGO8i5J0oI1e/XUR7tVbq1SdJtAXZveWfM/zZMkuwXh1RmaZl2R3KL5mgiaAABAi3GUd6ljRJCiQoP0bdExSdIl53TSnLF91CkiRKldo2v0jyVP02nz5s1TUlKSQkJClJaWps8++6zO/qtXr1ZaWppCQkLUvXt3vfDCC3afz58/X0OHDlV0dLSio6N16aWXav369U0+LwAAaJja8i79WFqub4uOKTjAT0+M7aO/T0q3rVUamRKnNfdcrEVTLtAz1/XToikXaM09F7d4wCR5WNC0ZMkSZWVl6f7779eWLVs0dOhQjRo1Svn5+Q775+XlafTo0Ro6dKi2bNmi++67T9OmTdPSpUttfVatWqXx48dr5cqVys3NVdeuXZWZmamDBw82+rwAAKBh6sq7VC0yNEDXpHWRYdg/cvP3M5RxVntd2S9BGWe1d0sJFUkyTNNsuaxQ9Rg4cKBSU1OVnZ1ta+vdu7fGjBmjOXPm1Oh/zz33aNmyZdq1a5etberUqdq2bZtyc3MdnsNqtSo6OlrPP/+8brjhhkad1xGLxaKoqCiVlJQoMjLSqe8AANBa5H53WOPnr6u336IpFyjjrPYtMKLTGvL722NmmsrLy7Vp0yZlZmbatWdmZmrt2rUOv5Obm1uj/4gRI7Rx40ZVVFQ4/M6JEydUUVGhdu3aNfq8AACgYTwx71JDecxC8OLiYlmtVsXExNi1x8TEqLCw0OF3CgsLHfavrKxUcXGx4uJqPu+89957lZCQoEsvvbTR55WksrIylZWV2d5bLHVvjQQAoLUqr6zSmj3FTvVtybxLDeUxQVO1M59jmqZZo62+/o7aJenJJ5/UokWLtGrVKoWE2N+Uhp53zpw5evjhh2v9HACA1sRRGgF/P0Ob849q5tLt+uaH0jq/7468Sw3lMUFThw4d5O/vX2N2p6ioqMYsULXY2FiH/QMCAtS+vf3z0Keeekp//vOf9fHHH+u8885r0nklaebMmZoxY4btvcViUZcuXeq+SAAAfJCjNAIxkcHqHRep1bt/lGlK7cKCNKZfvF7+fJ8MeUbepYbymDVNQUFBSktLU05Ojl17Tk6OBg0a5PA7GRkZNfqvWLFC6enpCgwMtLX9v//3//Too4/qww8/VHp6epPPK0nBwcGKjIy0ewEA0NrUlkbgB0uZVn1zOmAam9pZ/50xXA9dca5emJCq2Cj7pz2xUSHKnpDqljQCDeExM02SNGPGDE2cOFHp6enKyMjQSy+9pPz8fE2dOlXS6dmdgwcPauHChZJO75R7/vnnNWPGDE2ZMkW5ublasGCBFi1aZDvmk08+qQcffFD//Oc/1a1bN9uMUnh4uMLDw506LwAAqMmZNALt2gTqyWvOs80gjUyJ02XJsQ4f5Xk6jwqaxo0bp8OHD+uRRx5RQUGBUlJStHz5ciUmJkqSCgoK7HInJSUlafny5Zo+fbrmzp2r+Ph4Pfvssxo7dqytz7x581ReXq5rrrnG7lyzZs3S7NmznTovAACoaX3ekRozTGc6cqJC6/OO2KURqM675G08Kk+TNyNPEwCgtXlr8wHN+Ne2evs9c10/XdkvoQVG1HAN+f3tUTNNAADAc9S2I840TX3wVaGe/PAbp47jyWkEGoKgCQAA1OBoR1xcVIh+O6CL/vv1j9q6/ydJkp8hVdXyzMob0gg0BEETAACwU70j7sxYqKDklJ7O2SNJCg3015ShSUrqEGZ7ROeNaQQagqAJAADYOLMjrk2Qvz6580Jb6oDQIP8as1KxUSGadUWyx6cRaAiCJgAAYOPMjrgT5VblFR+3BU3enEagIQiaAABoRWpb3F3tywM/OXWcMwvremsagYYgaAIAoJWobXH3rCuSFRcVqnmrvtVHO35w6li+siOuIQiaAABoBepa3D319c12bSEBfjpVWeXwOL62I64hPKb2HAAAaB7OLO6WpKv6Jyhn+jD97bp+MvTzDrhqvrgjriEImgAA8HHOLO6WpN+kd1GPmAiNTIlTthcX1m0uPJ4DAMCL1bewW5L2Fh9z6li/XNzdWnbENQRBEwAAXqquhd0jU+K054dSvfz5Pr25ab9TxztzcXdr2BHXEARNAAB4ofoWdifHRWpngcXWHuBnqLKWeieteXF3QxA0AQDgZZxZ2L2zwCLDkDKTY3TzkO46fOyUbn1jiyTfL3fSXAiaAADwMs4u7P7rb/ppTP8E2/vsCUarKHfSXAiaAADwMPUt7v7BctKp4xhnTByxuLtpCJoAAPAgdS3uvuicTnr/ywI998m3Th3LUdZuFnc3HkETAAAeor7F3REhASo9VVnvcVjY3TwImgAA8ADOLO4uPVWp2MgQTcxIVExEsP705peSWNjdUgiaAADwAM4u7v5/156noT06SpLCQwJY2N2CCJoAAGgB9S3udjZr95Hj5bY/s7C7ZRE0AQDQzOpa3J3UIVyvfJ6npZsPOHUssna7D0ETAADNqL7F3b8U6G+owkrWbk9F0AQAQDNxZnG3JI1KidXNQ5L0YylZuz2Zn7sHAACAr3J2cfcNGd2U3q2dRvWJV/aEVMVG2T+Ci40KUfaEVBZ3uxkzTQAANEJ9C7sladP3R5w6VlHpz4EVi7s9F0ETAAANVNfC7szkWK3e/aNe+nSvcvcedup4LO72DgRNAAA0QH0Lu+OiQmzBlL+foUB/Q6cqqhwei8Xd3oU1TQAAOMmZhd0FJacUHhygKUOT9OndF+lv4/rJ0M+LuauxuNv7MNMEAICTnF3Y/cx1/XRJ7xhJUkLbUGVPSCVztw8gaAIAwEm/XLBdl2Nl9kV1WdztGwiaAAD4n7p2xH1bdExvbmpc1m6Jxd2+gKAJAADVviPupsHdtHV/iZZ/VSCzniyVLOz2bQRNAIBWr64dcf+3/Gvb+8zkGKUlttXjH3wjiazdrQ1BEwCgVXNmR1xIoJ/eumWwkuMjJUmJ7cNY2N0KETQBAFo1Z3bEnaqoUsnJCtt7Fna3TgRNAACfVl+5ky/ynMvafebOORZ2tz4ETQAAn1VXuZO4qFD99ePdWvXNj04dy9GOOLQuBE0AAJ9UX7mTav5+hoL8/XSywurwOOyIQzXKqAAAfI4zi7sl6erUBH1y53D9dVxfSp2gXgRNAACf42y5k2vTuiixfZhGpsQpe0KqYqPsH8HFRoUoe0IqO+IgicdzAAAf5Gy5k1/2Y0cc6kPQBADwOnXtiCs5WaFPvi5y6jhnLu5mRxzqQtAEAPAqte2Iu2/0OSooOaW5K7+zy6nkCIu70RgETQAAr1HXjrjbF221ve/RKVyX9u6k7NV7ZYhyJ3ANgiYAgFdwZkecnyHNufo8XZPWWf5+hvp2aUu5E7gMQRMAwCs4syOuypS6tmtjm0FicTdciaAJAOAVGrMjTmJxN1yHPE0AAI9XcrJCOTt/cKov5U7QXJhpAgC4XW0pBCqsVfrnF/n628e7dfQEO+LgXgRNAAC3cpRCIDYyWFf1T9BHO3/Q3h+PS5LO7hSuEefGaO7K79gRB7cgaAIAuE1tKQQKLWXKXr1XktQ+LEhZl/XU+PO7KMDfT30SotgRB7cgaAIAuIUzKQTCgv313zuHq22bIFsbO+LgLgRNAAC3cCaFwPEyq3YVlNbY/caOOLgDu+cAAG7xg+WkU/2cTTUANDdmmgAAzaKuorrfFpXq5TX7nDoOKQTgKQiaAAAuV1tR3btH9NJXhyx6be0+VVbVtZqJFALwPARNAACXqquo7vR/bbO9v7R3jIb1bK9Z7+6URAoBeD6CJgCAyzizI87fz9Dfb0jXRed0knT68RspBOANCJoAAC7jzI44a5WpkEB/23tSCMBbEDQBAFyGorrwZaQcAAC4TICfc79W2BEHb8RMEwCgyUzT1OIN+/V/7++qsx874uDNCJoAAE5zlHvpwNETunfpduXuPSxJ6ta+jfYdPkFRXfgcgiYAgFMc5V6KCAnQqQqrKqymQgL9dFdmL904OEk5OwvZEQefQ9AEAKhXbbmXSk9VSpJ6xUZo/sR0dW3fRhI74uCbCJoAAHVyJvdSyYlyJUSH2rWxIw6+ht1zAIA6OZN7qdBSpvV5R1poRIB7EDQBAOr0g+WkU/2czdEEeCuCJgBArX6wnNLLn+9zqi+5l+DrWNMEAK2cozQC/n6Glm8v0H1vb9dPJyrq/D65l9BaeNxM07x585SUlKSQkBClpaXps88+q7P/6tWrlZaWppCQEHXv3l0vvPCC3ec7duzQ2LFj1a1bNxmGob/97W81jjF79mwZhmH3io2NdeVlAYBH+vCrAg154hONn79OdyzeqvHz12nQ4//VdS/l6tY3NuunExVKSYjUrCt6y9DPuZaqkXsJrYlHBU1LlixRVlaW7r//fm3ZskVDhw7VqFGjlJ+f77B/Xl6eRo8eraFDh2rLli267777NG3aNC1dutTW58SJE+revbsef/zxOgOhc889VwUFBbbX9u3bXX59AOBJqtMInLnI+wdLmdbtPSLDkG676Cy9dctg3Ti4u7InpCo2yv4RXGxUiLInpJJ7Ca2CYZpmXbtIW9TAgQOVmpqq7OxsW1vv3r01ZswYzZkzp0b/e+65R8uWLdOuXT+n7Z86daq2bdum3NzcGv27deumrKwsZWVl2bXPnj1b77zzjrZu3drosVssFkVFRamkpESRkZGNPg4AtARrlakhT3xS56649mFBWn//pXYzSLU9ygO8VUN+f3vMTFN5ebk2bdqkzMxMu/bMzEytXbvW4Xdyc3Nr9B8xYoQ2btyoioq6n8Gfac+ePYqPj1dSUpKuu+467d27t87+ZWVlslgsdi8A8BbOpBE4fLy8RhqB6txLV/ZLUMZZ7QmY0Kp4TNBUXFwsq9WqmJgYu/aYmBgVFhY6/E5hYaHD/pWVlSouLnb63AMHDtTChQv10Ucfaf78+SosLNSgQYN0+PDhWr8zZ84cRUVF2V5dunRx+nwA4G7OpgcgjQDwM48JmqoZhv2/WkzTrNFWX39H7XUZNWqUxo4dqz59+ujSSy/V+++/L0l67bXXav3OzJkzVVJSYnvt37/f6fMBgLu1DQ1yqh9pBICfeUzKgQ4dOsjf37/GrFJRUVGN2aRqsbGxDvsHBASoffvGp+4PCwtTnz59tGfPnlr7BAcHKzg4uNHnAAB3yT98Qk9+9HWdfUgjANTkMTNNQUFBSktLU05Ojl17Tk6OBg0a5PA7GRkZNfqvWLFC6enpCgwMbPRYysrKtGvXLsXFsRsEgG/5aEehfvXcZ9pxyKKwYH9JpBEAnOUxM02SNGPGDE2cOFHp6enKyMjQSy+9pPz8fE2dOlXS6UdiBw8e1MKFCyWd3in3/PPPa8aMGZoyZYpyc3O1YMECLVq0yHbM8vJy7dy50/bngwcPauvWrQoPD9fZZ58tSbrrrrt0xRVXqGvXrioqKtJjjz0mi8WiSZMmtfDfAAA0naMdblWmqSc//FrzP8uTJKUlRuu58f315YGf9PB7O+0WhcdGhWjWFcmkEQDO4FFB07hx43T48GE98sgjKigoUEpKipYvX67ExERJUkFBgV3OpqSkJC1fvlzTp0/X3LlzFR8fr2effVZjx4619Tl06JD69+9ve//UU0/pqaee0vDhw7Vq1SpJ0oEDBzR+/HgVFxerY8eOuuCCC7Ru3TrbeQHAW3z4VUGNIKhTRLAiQgL03Y/HJUm/G5Kke0ado0B/P8W3DdVlybGkEQCc4FF5mrwZeZoAuFt1ssrafqiHBvrrr+P6aWQKFQ+Aal6ZpwkA0HjWKlMPv7ez1oBJksKD/XVZsuONNQDqR9AEAD7AmWSVPx6rmawSgPMImgDAB5CsEmh+BE0A4AM6hjuXN45klUDjedTuOQBAw5WcqNDLn+fV2YdklUDTETQBgBfbfqBEt/5zk/YfOakAP0OVVTWXgpOsEnANgiYA8AJnJqw8v1u0Fm3Yr0ff26lya5W6tAtV9vVpOnD0BMkqgWZC0AQAHs5RwsqQQD+dqqiSJF2WHKOnru2rqNBApSREkawSaCYETQDgwWpLWFkdMF2dmqCnr+0rw/g5KPL3M5RxVuOLlgNwjN1zAOChnElYmfvdYTlYxgSgGRA0AYCHciZhZUHJKRJWAi2EoAkAPBQJKwHPQtAEAB7KcrLSqX4krARaBgvBAcADLdmQr0f/s7POPiSsBFoWM00A4EHKK6t0/9vbdc/S7Sq3VqlvlyhJPyeorEbCSqDlMdMEAG5yZsLKbu3b6I+LtmjT90dlGNKdl/XUrReerRU7C0lYCXgAgiYAcANHCSv9DKnKlCJDAvTM+P66qFcnSdLIlDgSVgIegKAJAFpYbQkrq/Mt3TWipy1gqkbCSsD9WNMEAC3ImYSV2av2ykrGSsDjNGimadmyZQ0+wWWXXabQ0NAGfw8AfFFDElYyswR4lgYFTWPGjGnQwQ3D0J49e9S9e/cGfQ8AfBUJKwHv1eDHc4WFhaqqqnLq1aZNm+YYMwB4re+LTzjVj4SVgOdp0EzTpEmTGvSobcKECYqMjGzwoADA15imqXmrvtNfPt5dZz8SVgKeyzBNk9WGLmCxWBQVFaWSkhICRQB2TlVYdfebX2rZtkOSpIt6ddTKb36UIdktCK9OIJA9IZX8S0ALacjvb1IOAICLnJmsckBSOxUfK9PvF27UtgMlCvAz9MiVKfrtwK4O8zSRsBLwbA0KmpKSkmQYDU+mlpWVpWnTpjX4ewDgLRwFQR3Cg1RZZeqnExVq2yZQ2den2XbEkbAS8D4NCppeffXVRp2kW7dujfoeAHiD2pJVFh8rlyTFRYVo8e8vUGL7MLvPSVgJeJcGBU3Dhw9vrnEAgFdyJllllWmqczS7iQFv57KM4GVlZa46FAB4DWeSVf5gKdP6vCMtNCIAzcVlQdOgQYNqtO3eXffWWgDwdiSrBFqPJu+e+89//qOvv/5ax48f16FDhxQfH2/77Nprr9W2bduaegoA8FjOJqEkWSXg/ZocNJ177rnKz89XUVGRxo8fr/3796tz586Kj4+Xv7+/K8YIAB7LcqqiRr6lXyJZJeA7mhw0JSUl6dZbb1VKSoqGDRsmSTp48KDy8vKUkpLS5AECgCcyTVPzP9urOR98XWfAJEmzrkgmlQDgAxoVNFVUVGjkyJEKCwvTu+++K8MwbAGTJCUkJCghIcFlgwQAd3GUsNJaZerBd77Sko37JUnXD+yqQWe112Pv7yJZJeDDGhU0BQYGavPmzUpMTGxUsksA8AaOElbGRAQrqk2gdv9wTH6G9ODlyZo8qJsMw9DIlDiSVQI+rNGP56666ir961//0oEDB9S5c2dXjgkA3K62hJU/lJbph9IyhQT6Kfv6NF10TifbZySrBHxbo4OmDh06qLKyUqmpqRo7dqxiY2PtZp0eeughlwwQAFqaMwkrw4MDNKxnxxYbEwD3M0zTrOvnQq38/PxkGIZM03T4iM5qtTZ5cN6kIVWSAXi23O8Oa/z8dfX2WzTlAmaWAC/XkN/fjZ5p6tq1K+uZAPgkElYCcKTRQdO+fftcOAwA8BwkrATgiMvKqACAr+gdF6Eg/9p/PBqS4khYCbQ6BE0A8AsFJSd13UvrVG6tcvg5CSuB1ougCQD+Z1eBRVfNXauvC0vVMSJY940+R3FR9o/gYqNClD0hlYSVQCvU5DIqkrR582alpKQoKCjIFYcDgGblKMv3ur2HNfUfm1RaVqmzO4Xr1RvPV+foNrp5SHcSVgKQ5KKg6fzzz9euXbvUs2dPVxwOAJqNoyzfUaGBKj1VoSpTGpjUTi9NTFdUm0BJJKwE8DOXBE2NTPUEAC2qtizfJScrJEnpidFaePMABQf4t/zgAHg81jQBaBWcyfJ94OgJBfjxYxGAY/x0ANAqrM87YvdIzpFCS5nW5x1poREB8DYETQBaBbJ8A2gqgiYArQJZvgE0FUETgFahc3RonakCyPINoD4ETQB83t4fj2nci7myVjleBk6WbwDOcEnQNGvWLHXo0MEVhwIAl9pxqETXvpCrQyWn1L1jmP58VQpZvgE0imGSZMklLBaLoqKiVFJSosjISHcPB4CkjfuO6MZXN6j0VKVSEiL12o0D1D482GFGcGaYgNapIb+/XZLcEgDc7cxA6FSFVbe8sUmnKqo0oFs7/X1yuiJDyPINoPEImgB4PUelUapd2Kujsq9PU2gQWb4BNA1BEwCvVltplGrXpCYQMAFwCZcETT/99JM++ugjHTx4UIZhKC4uTiNGjFB0dLQrDg8ADjlTGuX/ln+tUX3iWbMEoMmavHtuwYIFGjBggNatW6eqqipZrVatW7dOF1xwgRYsWOCKMQKAQ86URikoOUVpFAAu0eSZpieffFKbN29WeHi4Xfujjz6qtLQ03XzzzU09BQA4RGkUAC2pyTNNhmHo2LFjNdqPHTsmw2A6HEDzoTQKgJbU5Jmmp556SsOHD1dKSooSEhIkSQcOHNCOHTv09NNPN3mAAFCb3O+K6/zc0OnElZRGAeAKTQ6aLr/8co0aNUrr16/XoUOHZJqmEhISNGDAAPn7s2MFgOuZpqm/frxHz37yra3NkOwWhFMaBYCrNSpoqqio0MiRIxUWFqZ3331X/v7+ysjIcPXYAKAG0zT1l5zdeu5/AdP9o3urS7vQGnmaYqNCNOuKZEqjAHCZRgVNgYGB2rx5sxITE1m3BKDFmKapp1fs1vMrTwdMD/yqt343tLsk6bLkWEqjAGhWjX48d9VVV+lf//qXDhw4oM6dO7tyTABQoyzK+d2i9Zec3Zq36jtJ0oOXJ+vmIUm2/pRGAdDcGh00dejQQZWVlUpNTdXYsWMVGxtrN+v00EMPuWSAAFofR2VRwoL9dbzMKkl66PJk3fSLgAkAWoJhmmZdyXRr5efnJ8MwZJqmw0d0Vqu1yYPzJg2pkgygdvWVRflNemc9eU3fFh0TAN/VkN/fjZ5p6tq1K+uZALiUM2VRPttTLGuVyXolAC2u0UHTvn37XDgMAGhYWRTWLwFoaU3OCA4ArkJZFACezOOCpnnz5ikpKUkhISFKS0vTZ599Vmf/1atXKy0tTSEhIerevbteeOEFu8937NihsWPHqlu3bjIMQ3/7299ccl4ArkdZFACerEGP55KSkhq1jikrK0vTpk2rt9+SJUuUlZWlefPmafDgwXrxxRc1atQo7dy5U127dq3RPy8vT6NHj9aUKVP0+uuv6/PPP9ett96qjh07auzYsZKkEydOqHv37rr22ms1ffp0l5wXQPMYkNROUaGBKjlZ4fBzyqIAcKcG7Z5bvXp1o07SrVs3JSYm1ttv4MCBSk1NVXZ2tq2td+/eGjNmjObMmVOj/z333KNly5Zp165dtrapU6dq27Ztys3NdTiOrKwsZWVlNem8jrB7Dmi6j3YU6pbXN6nKwU+l6n+uZU9IJcs3AJdptt1zw4cPb9LA6lJeXq5Nmzbp3nvvtWvPzMzU2rVrHX4nNzdXmZmZdm0jRozQggULVFFRocDAwGY5rySVlZWprKzM9t5isdR7LgC1+/zbYt3+zy2qMqULurfTvuLjKrT8/P8YZVEAuFuDgqYvv/xSKSkp8vNzbinUjh071KtXLwUE1H+a4uJiWa1WxcTE2LXHxMSosLDQ4XcKCwsd9q+srFRxcbHi4ur/4dqY80rSnDlz9PDDD9d7fAD125x/VFMWblS5tUojz43V87/tL8MwKIsCwKM0aCF4//79dfjwYaf7Z2RkKD8/v0EDOnPNVG3JM+vq76jd1eedOXOmSkpKbK/9+/c36HwATttVYNHkl9frRLlVQ3t00DPj+ynA389WFuXKfgnKOKs9ARMAt2vQTJNpmnrwwQfVpk0bp/qXl5c7fewOHTrI39+/xuxOUVFRjVmgarGxsQ77BwQEqH1753K4NOa8khQcHKzg4GCnzgHgtDPryXWMCNbEBetlOVWptMRovTgxTcEB/u4eJgA41KCgadiwYfrmm2+c7p+RkaHQ0FCn+gYFBSktLU05OTm66qqrbO05OTm68soraz3+e++9Z9e2YsUKpaenO7WeqbHnBdBwjurJ+RuS1ZSS4yL18uTz1Sao0fl2AaDZNegn1KpVq5ppGKfNmDFDEydOVHp6ujIyMvTSSy8pPz9fU6dOlXT6kdjBgwe1cOFCSad3yj3//POaMWOGpkyZotzcXC1YsECLFi2yHbO8vFw7d+60/fngwYPaunWrwsPDdfbZZzt1XgBNU1s9Oev/GiYPTlRUqHP/0AEAd/Gof9aNGzdOhw8f1iOPPKKCggKlpKRo+fLltnQFBQUFdmukkpKStHz5ck2fPl1z585VfHy8nn32WVuOJkk6dOiQ+vfvb3v/1FNP6amnntLw4cNtQWB95wXQeM7Uk/trzh6NTe3CuiUAHq1BeZp+qaqqSq+88or++9//6ocfftAvD2MYhv773/+6bJDegDxNgGO53x3W+Pnr6u23aMoF1JMD0OKaLU/TL82YMUPPPfecJPsda/XtOgPQulBPDoCvaHTQtGjRIpmmqfj4eCUlJTmViwlA60M9OQC+otGRjtVqVefOnbVnzx623gOo1YCkdgoPDtCxskqHn1NPDoC3aFByy1+67rrrdPLkSVVUOC6sCQCStHTzgToDJkmadUUyi8ABeLxGzzSFh4fLYrGoX79++vWvf622bdvaff7QQw81dWwAvNzKb4o0863tkqQR58boywMldnmaqCcHwJs0evecn59fnQu/rVZrkwfnTdg9B9jbtv8nXffSOp2ssOrq/gl6+jd9VWWKenIAPEqL7J7r2rUru+QAOLSv+LhuenWDTlacrif3xDXnyTAM+RsirQAAr9XooGnfvn0uHAYAX/FjaZlueHm9Dh8vV5+EKGVPSFOgf6OXTwKAxyBPAIAm+WUR3siQQD2d843yj5xQ13Zt9PLk8xUezI8ZAL6hST/NysvL9fnnn+vQoUM11jDdcMMNTRoYAM/nqAivJIUHB+i1mwaoYwTpSAD4jkYHTXv27NFll12m/fv31/jMMAyCJsDH1VaEV5KOlVXqm0KLkjqEtfi4AKC5NHqhwb333qv8/HyZpunwBcB31VeE15D08Hs7Za3iZwEA39HooGnNmjUKCAhQTk6OJKl///5atGiROnToYGsD4JvW5x2p8Ujul0xJBSWntD7vSMsNCgCaWaODpqNHj6p379665JJLZBiGAgMDNW7cOMXGxurPf/6zK8cIwMNQhBdAa9ToNU0RERGqqqqSdDo7+Ndff60vvvhC+fn5+u6771w2QACehyK8AFqjRs80de3aVd9//72sVqv69Omj0tJSDRo0SKWlpYqLoyQC4MvaBPmrrtS2hqQ4ivAC8DGNDpquv/56DR8+XLt379b999+vwMBAmaYpPz8/zZ4924VDBOBJ9hUf182vbahzEbhEEV4AvqfRtefOlJeXpy1btujcc89Vr169XHFIr0LtObQGRaWndE12rvKPnFByXKSmDE3Skx99Y7coPI4ivAC8SIvUnpOk4uJiPffcc1q3bp0SExM1bdo0ffHFFwoNDVXXrl2bcmgAHqb0VIUmv7zBlu371ZvOV6eIEP26XwJFeAG0Ck2qPTd48GAVFhZKkgYOHKiSkhJNnjxZd911l5588kmXDRKAe5VVWvX7hZu0s8CiDuFBWnjTANsib38/gyK8AFqFRgdNd999twoKCtS5c2cdOHBAkjR48GBFRkaSpwnwcr+sJ9chLFivf/G9cvceVliQv169cYC6kekbQCvU6KDp448/VocOHbRr1y5FRETY2hMTE7Vv3z5XjA2AG9RWT87fz9BLN6QrJSHKTSMDAPdqdNB08uRJ9ejRQ2Fh9v/iPHbsmMrKypo8MAAtr656ctYqU6WnKlp8TADgKRqdcuCss87Sjh079Prrr0uSysrK9NxzzykvL089e/Z02QABtAzqyQFA3RodNE2ZMkWmaWrSpEkyDENbt25VVlaWDMPQTTfd5MoxAmgB1JMDgLo1OmiaNm2apk6dKkkyTVPV6Z6mTJmiadOmuWZ0AFoM9eQAoG6NXtNkGIbmzZunu+++Wxs3bpRpmkpPT1dSUpIrxweghVBPDgDq1qTkllarVeXl5YqJiZFpmtq/f7/2798vSRo2bJhLBgigZcRFhcjPkGpbsmRIiqWeHIBWrNFB09q1a/Xb3/7WFiT9kmEYqqysbNLAALSc4mNluvHVDXUGTBL15AC0bo1e03TrrbcqPz/ftp7pzBcA73CsrFI3vrJBecXHldA2VI9fnaK4KPtHcLFRIcqekEo9OQCtWqNnmr799ltFR0fr3//+t7p37y7D4F+fgLcpr6zS1H9s0vaDJWoXFqR/3DxA3TuG69r0rtSTA4AzNDpoGj58uLZv365hw4YpIKBJS6MAuEFVlak7/71Na74tVpsgf70y+Xx17xguiXpyAOBIo6OdBQsW6MILL1RqaqoyMzMVGRlp9/lDDz3U5MEBaB6maeqR/+zUe9sOKdDf0AsT0tS3S1t3DwsAPFqjg6YPP/xQ3333naqqqrRjx44anxM0AZ7jlwV4O0WEaMO+I3p17T5J0lPX9tWwnh3dO0AA8AKNDpoeeOABWa1WV44FQDOorQCvdHo33JX9EtwwKgDwPo3ePXfs2DHFxcVpz549qqioUFVVld0LgPtVF+CtrTzKmbvkAAC1a3TQdPPNN6uyslKdOnWSv7+/K8cEwAXqK8ArUYAXABqi0Y/nfvzxR1ksFvXo0UNDhgyxWwhuGIYWLFjgkgECaJz6CvBKPxfgZaccANSv0UHT66+/LsMwVFRUpLffftvWbpomQRPgASjACwCu1eigqWvXriS0BDwYBXgBwLUaHTTt27fPhcMA4Gq9YiMU4GeospY1SxTgBYCGafRCcACe62S5Vb9fuLHOgEmiAC8ANARBE+BjKqxVuvWNTdr4/VFFhgTowV/1pgAvALgAReMAH1JVZepP/96mld/8qJBAP708+Xyld2unyYOTKMALAE1E0AT4iOp6cu9sPaQAP0PZ16cpvdvp9UoU4AWApiNoArzUmfXk1ufZ15O76JxO7h0gAPgYgibAC9VXT25Mf+rJAYCrETQBXqa6nlxtxU+oJwcAzYPdc4AXoZ4cALgPQRPgRRpSTw4A4FoETYAXoZ4cALgPQRPgRagnBwDuQ9AEeJHO0aGqKyelodMLwaknBwCuR9AEeIkfS8t0w8vrVdsab+rJAUDzImgCvEDJiQpNXPCF8oqPK6FtqP58VQr15ACghZGnCfBwx8sqNfnV9fq6sFQdwoP1+u8GKqlDmMad35V6cgDQggiaAA92qsKq3/9jo7bk/6So0EC9/rsBSuoQJol6cgDQ0giaAA/yy3py7cKC9Nra7/X5t4fVJshfr954vs6JjXT3EAGg1SJoAjxEbfXkAvwM/X1Suvp3jXbTyAAAEkET4BHqqidXWWXKcrKixccEALDH7jnAzeqrJ2eIenIA4AkImgA3q6+enCnqyQGAJyBoAtyMenIA4B0ImgA3o54cAHgHFoIDbrb94E91fm7odLZv6skBgHsx0wS40ctr8vTn5V/b3p+Zz5t6cgDgOQiaADdZmLtPj/xnpyTp9ovPVvb1/RVLPTkA8Fg8ngPc4PV13+uhd3dIkm658CzNuKynDMNQ5rlx1JMDAA9F0AQ0o1+WRakOgv61cb8eeOcrSdLvh3XX3SN6yTBOB0bUkwMAz0XQBDQTR2VRokIDVHKyUpJ085AkzRx1ji1gAgB4No9b0zRv3jwlJSUpJCREaWlp+uyzz+rsv3r1aqWlpSkkJETdu3fXCy+8UKPP0qVLlZycrODgYCUnJ+vtt9+2+3z27NkyDMPuFRsb69LrQutSXRblzKSV1QHTRb066oFf9SZgAgAv4lFB05IlS5SVlaX7779fW7Zs0dChQzVq1Cjl5+c77J+Xl6fRo0dr6NCh2rJli+677z5NmzZNS5cutfXJzc3VuHHjNHHiRG3btk0TJ07Ub37zG33xxRd2xzr33HNVUFBge23fvr1ZrxW+q76yKJK0q8AiqqIAgHcxTNP0mB/dAwcOVGpqqrKzs21tvXv31pgxYzRnzpwa/e+55x4tW7ZMu3btsrVNnTpV27ZtU25uriRp3Lhxslgs+uCDD2x9Ro4cqejoaC1atEjS6Zmmd955R1u3bm302C0Wi6KiolRSUqLIyMhGHwfeL/e7wxo/f129/RZNuYD1SwDgZg35/e0xM03l5eXatGmTMjMz7dozMzO1du1ah9/Jzc2t0X/EiBHauHGjKioq6uxz5jH37Nmj+Ph4JSUl6brrrtPevXvrHG9ZWZksFovdC5AoiwIAvspjgqbi4mJZrVbFxMTYtcfExKiwsNDhdwoLCx32r6ysVHFxcZ19fnnMgQMHauHChfroo480f/58FRYWatCgQTp8+HCt450zZ46ioqJsry5dujToeuG7KIsCAL7JY4KmamcujDVNs87Fso76n9le3zFHjRqlsWPHqk+fPrr00kv1/vvvS5Jee+21Ws87c+ZMlZSU2F779++v58rQWgxIaqeo0No3phqS4iiLAgBex2NSDnTo0EH+/v41ZpWKiopqzBRVi42Nddg/ICBA7du3r7NPbceUpLCwMPXp00d79uyptU9wcLCCg4PrvCa0Tks3HZDlf7vkzkRZFADwXh4z0xQUFKS0tDTl5OTYtefk5GjQoEEOv5ORkVGj/4oVK5Senq7AwMA6+9R2TOn0eqVdu3YpLo7SFWiYxevzdffSL2VKGt6zo2Ij7QNryqIAgPfymJkmSZoxY4YmTpyo9PR0ZWRk6KWXXlJ+fr6mTp0q6fQjsYMHD2rhwoWSTu+Ue/755zVjxgxNmTJFubm5WrBggW1XnCTdcccdGjZsmJ544gldeeWVevfdd/Xxxx9rzZo1tj533XWXrrjiCnXt2lVFRUV67LHHZLFYNGnSpJb9C4BX++cX+brv7dOpKiYP6qZZVySryhRlUQDAR3hU0DRu3DgdPnxYjzzyiAoKCpSSkqLly5crMTFRklRQUGCXsykpKUnLly/X9OnTNXfuXMXHx+vZZ5/V2LFjbX0GDRqkxYsX64EHHtCDDz6os846S0uWLNHAgQNtfQ4cOKDx48eruLhYHTt21AUXXKB169bZzgv8kqPSKP9cn68H/1ca5abBSXrw8tOJK/0NkVYAAHyER+Vp8mbkaWodHJVGiQwJkOXU6TVMvxuSpPvJ9A0AXqMhv789aqYJ8GTVpVHO/FdGdcB0WXInAiYA8GEesxAc8GTOlEbZfqCE0igA4MMImgAnrM87UqP47pkKLWVan3ekhUYEAGhpBE2AEyiNAgAgaAKcQGkUAABBE+CE9MRotQnyr/VzSqMAgO8jaALqUWmt0p/e3KYT5VaHn1MaBQBaB4ImoA5llVbd+sZmvbP1kAL8DP1uSDfFRdk/gqM0CgC0DuRpAmpxorxSf/jHJn22p1hBAX6a99tUXZoco5mjkymNAgCtEEEToJqlUXrHRWjKwo3asO+o2gT5a/4N6Rp8dgdJkr+fQWkUAGiFCJrQ6jkqjRLob6jCaioiJECv3jhAaYnRbhwhAMATEDShVautNEqF9XTL7RefTcAEAJDEQnC0Ys6URnnl832yUhsFACCCJrRizpRGKSg5RWkUAIAkgia0YpRGAQA0BEETWi1KowAAGoKF4GiVKq1VWr69oM4+hk4nrqQ0CgBAImhCK1R6qkK3L9qiVd/8WGsfSqMAAM5E0IRW5eBPJ3Xzqxv0dWGpQgL99Nff9JNhqEaeptioEM26IpnSKAAAG4Im+Kwzs3yHBPppysJNKj5Wpo4Rwfr7Denq26WtJOmy5FhKowAA6kTQBJ/kKMt3tXNiI7Rg8vlKaBtqa6M0CgCgPgRN8Dm1ZfmuNnV4d7uACQAAZ5ByAD7FmSzfT3z4DVm+AQANRtAEn0KWbwBAcyFogk8hyzcAoLkQNMFnmKapbftLnOpLlm8AQEOxEBw+4VSFVQ+9+5X+tfFAnf3I8g0AaCxmmuD1Dv10UuNezNW/Nh6QnyFd1T9e0s9ZvauR5RsA0BTMNMFrnJmsckBSO23Yd0S3vbFZh4+Xq22bQD03vr+G9uioEefGkuUbAOBSBE3wCo6SVUaGBOhYWaWqTKl3XKRempimLu3aSJJGpsSR5RsA4FIETfB4tSWrtJyqlCQN6Bat124aqNAgf7vPyfINAHAl1jTBozmTrDL/yAkFBfCfMgCgefGbBh7NmWSVhZYyklUCAJodQRM8GskqAQCegqAJHq1NoHPL7khWCQBobiwEh8fafqBEs9/bUWcfklUCAFoKQRPcylHuJX8/Q4vX5+uhZTtUXlmljuHB+vFYmQzJbkE4ySoBAC2JoAlu4yj3UmxksLp3DNfa7w5Lki7t3UlP/6afcr8rJlklAMCtDNM069rNDSdZLBZFRUWppKREkZGR7h6Ox6st91I1w5DuyuylW4afJb//zSLVNisFAEBjNeT3NzNNaHHO5F6KbhOoqb8ImCSSVQIA3Ivdc2hxzuReOnK8gtxLAACPQtCEFkfuJQCANyJoQotzNqcSuZcAAJ6EoAkt7vD/0gfUxpAUR+4lAICHYSE4Woy1ytRTK75R9qrvau1D7iUAgKciaILLOUoNUHqqQrcv2qLP9hRLkqYMTVK/Lm312Pu7yL0EAPAKBE1wKUcJKzuEB0mSio+VKyTQT09e01e/7hsvSRqZEkfuJQCAVyBogsvUlrCy+Fi5pNPB08KbBio5/ufkYeReAgB4CxaCwyWcSVjpbxjqFRvRYmMCAMCVCJrgEs4krPyhtIyElQAAr0XQBJcgYSUAwNcRNMElSFgJAPB1LARHk5VXVmn59oI6+xg6nU6AhJUAAG9F0IQmKSo9pdve2KwN+47W2oeElQAAX0DQBKc4Sli5df9PuuX1TSoqLVNEcID+Oq6fKquqauRpImElAMAXEDShXo4SVkaGBOh4uVXWKlM9OoXrxYlp6t4xXJJ0WXIsCSsBAD6HoAl1qi1hpeVUpSSpf9e2+sfNAxUe/PN/SiSsBAD4InbPoVbOJKws+OmkQgP9W2xMAAC4C0ETauVMwspCCwkrAQCtA0ETakXCSgAAfkbQhFqVVVQ51Y+ElQCA1oCF4KihqsrUq2v36fEPv66zHwkrAQCtCUFTK+Uo75K/n6Eiyynd9eaX+nT3j5KklPhIfXXIIkOyWxBOwkoAQGtD0NQKOcq7FBcVoiv7xWvJhv06eqJCwQF+uv9XvTXxgkR9tKOQhJUAgFbPME2zrh3lcJLFYlFUVJRKSkoUGRnp7uHUqra8S7+UHBepZ67rpx4xEba22mamAADwZg35/c1MUyviTN6lsCB/vXlLhtoE2f+nQcJKAEBrx+65VsSZvEvHy63atr+khUYEAID3IGhqRci7BABA4/F4zofUte7o8LEy/XdXkVPHIe8SAAA1ETT5iNp2xM24rIe+/fG4/pH7vU6UW+s8BnmXAACoHUGTh3Nm11ptO+IKSk7pT29ut71PSYjUsB4dNG/VXvIuAQDQQB63pmnevHlKSkpSSEiI0tLS9Nlnn9XZf/Xq1UpLS1NISIi6d++uF154oUafpUuXKjk5WcHBwUpOTtbbb7/d5PO2hA+/KtCQJz7R+PnrdMfirRo/f52GPPGJPvyqwNbHmR1xgf6G5t+Qpvf+OER3j+ytFyakKjbK/hFcbFSIsiekkncJAIBaeNRM05IlS5SVlaV58+Zp8ODBevHFFzVq1Cjt3LlTXbt2rdE/Ly9Po0eP1pQpU/T666/r888/16233qqOHTtq7NixkqTc3FyNGzdOjz76qK666iq9/fbb+s1vfqM1a9Zo4MCBjTpvS6ht9qiw5JRueX2znh3fT2d1jNB72w7VuyOuwmoqPDhQhnF6BmlkSpwuS44l7xIAAA3gUcktBw4cqNTUVGVnZ9vaevfurTFjxmjOnDk1+t9zzz1atmyZdu3aZWubOnWqtm3bptzcXEnSuHHjZLFY9MEHH9j6jBw5UtHR0Vq0aFGjzuuIK5NbWqtMDXnik3qDoYZ45rp+urJfgsuOBwCAL2jI72+PeTxXXl6uTZs2KTMz0649MzNTa9eudfid3NzcGv1HjBihjRs3qqKios4+1cdszHklqaysTBaLxe7lKs7kU5JOJ6I8Jzai3n4SO+IAAGgqjwmaiouLZbVaFRMTY9ceExOjwsJCh98pLCx02L+yslLFxcV19qk+ZmPOK0lz5sxRVFSU7dWlSxfnLtQJzuZJ+r+rUvT+tKGKiwpRbQ/WDJ3eRceOOAAAmsZjgqZq1etuqpmmWaOtvv5ntjtzzIaed+bMmSopKbG99u/fX2vfhnJ2VigmMlT+foZmXZEsSTUCJ3bEAQDgOh4TNHXo0EH+/v41ZneKiopqzAJVi42Nddg/ICBA7du3r7NP9TEbc15JCg4OVmRkpN3LVQYktWvQ7NHIlDhlsyMOAIBm5TFBU1BQkNLS0pSTk2PXnpOTo0GDBjn8TkZGRo3+K1asUHp6ugIDA+vsU33Mxpy3uTVm9mhkSpzW3HOxFk25QM9c10+LplygNfdcTMAEAICLeFTKgRkzZmjixIlKT09XRkaGXnrpJeXn52vq1KmSTj8SO3jwoBYuXCjp9E65559/XjNmzNCUKVOUm5urBQsW2HbFSdIdd9yhYcOG6YknntCVV16pd999Vx9//LHWrFnj9HndoXr26Mws37FRIZp1RbLDYMjfz1DGWe1bcpgAALQepoeZO3eumZiYaAYFBZmpqanm6tWrbZ9NmjTJHD58uF3/VatWmf379zeDgoLMbt26mdnZ2TWO+e9//9vs1auXGRgYaJ5zzjnm0qVLG3ReZ5SUlJiSzJKSkgZ9rz6V1ipz7bfF5jtbDphrvy02K61VLj0+AACtWUN+f3tUniZv5so8TQAAoGV4ZZ4mAAAAT0bQBAAA4ASCJgAAACcQNAEAADiBoAkAAMAJBE0AAABOIGgCAABwAkETAACAEwiaAAAAnOBRtee8WXVidYvF4uaRAAAAZ1X/3namQApBk4uUlpZKkrp06eLmkQAAgIYqLS1VVFRUnX2oPeciVVVVOnTokCIiImQYhkuPbbFY1KVLF+3fv98n69pxfd7P16/R169P8v1r5Pq8X3Ndo2maKi0tVXx8vPz86l61xEyTi/j5+alz587Neo7IyEif/Z9B4vp8ga9fo69fn+T718j1eb/muMb6ZpiqsRAcAADACQRNAAAATiBo8gLBwcGaNWuWgoOD3T2UZsH1eT9fv0Zfvz7J96+R6/N+nnCNLAQHAABwAjNNAAAATiBoAgAAcAJBEwAAgBMImgAAAJxA0OTh5s2bp6SkJIWEhCgtLU2fffaZu4fkMrNnz5ZhGHav2NhYdw+r0T799FNdccUVio+Pl2EYeuedd+w+N01Ts2fPVnx8vEJDQ3XhhRdqx44d7hlsI9R3fZMnT65xPy+44AL3DLYR5syZo/PPP18RERHq1KmTxowZo2+++cauj7ffQ2eu0ZvvY3Z2ts477zxb8sOMjAx98MEHts+9/f7Vd33efO8cmTNnjgzDUFZWlq3N3feQoMmDLVmyRFlZWbr//vu1ZcsWDR06VKNGjVJ+fr67h+Yy5557rgoKCmyv7du3u3tIjXb8+HH17dtXzz//vMPPn3zySf3lL3/R888/rw0bNig2NlaXXXaZrW6hp6vv+iRp5MiRdvdz+fLlLTjCplm9erVuu+02rVu3Tjk5OaqsrFRmZqaOHz9u6+Pt99CZa5S89z527txZjz/+uDZu3KiNGzfq4osv1pVXXmn7pert96++65O8996dacOGDXrppZd03nnn2bW7/R6a8FgDBgwwp06datd2zjnnmPfee6+bRuRas2bNMvv27evuYTQLSebbb79te19VVWXGxsaajz/+uK3t1KlTZlRUlPnCCy+4YYRNc+b1maZpTpo0ybzyyivdMp7mUFRUZEoyV69ebZqm791D06x5jabpe/cxOjra/Pvf/+6T9880f74+0/Sde1daWmr26NHDzMnJMYcPH27ecccdpml6xv+DzDR5qPLycm3atEmZmZl27ZmZmVq7dq2bRuV6e/bsUXx8vJKSknTddddp79697h5Ss8jLy1NhYaHd/QwODtbw4cN96n6uWrVKnTp1Us+ePTVlyhQVFRW5e0iNVlJSIklq166dJN+8h2deYzVfuI9Wq1WLFy/W8ePHlZGR4XP378zrq+YL9+62227Tr371K1166aV27Z5wDynY66GKi4tltVoVExNj1x4TE6PCwkI3jcq1Bg4cqIULF6pnz5764Ycf9Nhjj2nQoEHasWOH2rdv7+7huVT1PXN0P7///nt3DMnlRo0apWuvvVaJiYnKy8vTgw8+qIsvvlibNm3yuizFpmlqxowZGjJkiFJSUiT53j10dI2S99/H7du3KyMjQ6dOnVJ4eLjefvttJScn236pevv9q+36JO+/d5K0ePFibd68WRs2bKjxmSf8P0jQ5OEMw7B7b5pmjTZvNWrUKNuf+/Tpo4yMDJ111ll67bXXNGPGDDeOrPn48v0cN26c7c8pKSlKT09XYmKi3n//fV199dVuHFnD/fGPf9SXX36pNWvW1PjMV+5hbdfo7fexV69e2rp1q3766SctXbpUkyZN0urVq22fe/v9q+36kpOTvf7e7d+/X3fccYdWrFihkJCQWvu58x7yeM5DdejQQf7+/jVmlYqKimpE2b4iLCxMffr00Z49e9w9FJer3hXYmu5nXFycEhMTve5+3n777Vq2bJlWrlypzp0729p96R7Wdo2OeNt9DAoK0tlnn6309HTNmTNHffv21TPPPOMz96+263PE2+7dpk2bVFRUpLS0NAUEBCggIECrV6/Ws88+q4CAANt9cuc9JGjyUEFBQUpLS1NOTo5de05OjgYNGuSmUTWvsrIy7dq1S3Fxce4eisslJSUpNjbW7n6Wl5dr9erVPns/Dx8+rP3793vN/TRNU3/84x/11ltv6ZNPPlFSUpLd575wD+u7Rke87T6eyTRNlZWV+cT9c6T6+hzxtnt3ySWXaPv27dq6davtlZ6eruuvv15bt25V9+7d3X8PW2S5ORpl8eLFZmBgoLlgwQJz586dZlZWlhkWFmbu27fP3UNziTvvvNNctWqVuXfvXnPdunXm5ZdfbkZERHjt9ZWWlppbtmwxt2zZYkoy//KXv5hbtmwxv//+e9M0TfPxxx83o6KizLfeesvcvn27OX78eDMuLs60WCxuHrlz6rq+0tJS88477zTXrl1r5uXlmStXrjQzMjLMhIQEr7m+W265xYyKijJXrVplFhQU2F4nTpyw9fH2e1jfNXr7fZw5c6b56aefmnl5eeaXX35p3nfffaafn5+5YsUK0zS9//7VdX3efu9q88vdc6bp/ntI0OTh5s6dayYmJppBQUFmamqq3dZgbzdu3DgzLi7ODAwMNOPj482rr77a3LFjh7uH1WgrV640JdV4TZo0yTTN09tlZ82aZcbGxprBwcHmsGHDzO3bt7t30A1Q1/WdOHHCzMzMNDt27GgGBgaaXbt2NSdNmmTm5+e7e9hOc3RtksxXXnnF1sfb72F91+jt9/Gmm26y/bzs2LGjeckll9gCJtP0/vtX1/V5+72rzZlBk7vvoWGaptkyc1oAAADeizVNAAAATiBoAgAAcAJBEwAAgBMImgAAAJxA0AQAAOAEgiYAAAAnEDQBAAA4gaAJgNdbtWqVDMPQTz/95O6hAPBhBE0AvM6FF16orKws2/tBgwapoKBAUVFRbhsTgRvg+wLcPQAAaKqgoCBbFXsAaC7MNAHwKpMnT9bq1av1zDPPyDAMGYahV1991W6W59VXX1Xbtm31n//8R7169VKbNm10zTXX6Pjx43rttdfUrVs3RUdH6/bbb5fVarUdu7y8XHfffbcSEhIUFhamgQMHatWqVbbPv//+e11xxRWKjo5WWFiYzj33XC1fvlz79u3TRRddJEmKjo6WYRiaPHmypNNV6J988kl1795doaGh6tu3r958803bMatnqN5//3317dtXISEhGjhwoLZv397sf5cAGoaZJgBe5ZlnntHu3buVkpKiRx55RJK0Y8eOGv1OnDihZ599VosXL1ZpaamuvvpqXX311Wrbtq2WL1+uvXv3auzYsRoyZIjGjRsnSbrxxhu1b98+LV68WPHx8Xr77bc1cuRIbd++XT169NBtt92m8vJyffrppwoLC9POnTsVHh6uLl26aOnSpRo7dqy++eYbRUZGKjQ0VJL0wAMP6K233lJ2drZ69OihTz/9VBMmTFDHjh01fPhw23j/9Kc/6ZlnnlFsbKzuu+8+/frXv9bu3bsVGBjYAn+rAJzSYqWBAcBFzqx8vnLlSlOSefToUdM0TfOVV14xJZnffvutrc8f/vAHs02bNmZpaamtbcSIEeYf/vAH0zRN89tvvzUNwzAPHjxod65LLrnEnDlzpmmaptmnTx9z9uzZDsd05hhM0zSPHTtmhoSEmGvXrrXre/PNN5vjx4+3+97ixYttnx8+fNgMDQ01lyxZ4uTfCICWwEwTAJ/Upk0bnXXWWbb3MTEx6tatm8LDw+3aioqKJEmbN2+WaZrq2bOn3XHKysrUvn17SdK0adN0yy23aMWKFbr00ks1duxYnXfeebWOYefOnTp16pQuu+wyu/by8nL179/fri0jI8P253bt2qlXr17atWtXA68aQHMiaALgk858rGUYhsO2qqoqSVJVVZX8/f21adMm+fv72/WrDrR+97vfacSIEXr//fe1YsUKzZkzR08//bRuv/12h2OoPvb777+vhIQEu8+Cg4PrvQbDMOrtA6DlEDQB8DpBQUF2C7hdoX///rJarSoqKtLQoUNr7delSxdNnTpVU6dO1cyZMzV//nzdfvvtCgoKkiS7cSUnJys4OFj5+fl265ccWbdunbp27SpJOnr0qHbv3q1zzjnHBVcGwFUImgB4nW7duumLL77Qvn37FB4ebpvRaYqePXvq+uuv1w033KCnn35a/fv3V3FxsT755BP16dNHo0ePVlZWlkaNGqWePXvq6NGj+uSTT9S7d29JUmJiogzD0H/+8x+NHj1aoaGhioiI0F133aXp06erqqpKQ4YMkcVi0dq1axUeHq5JkybZzv/II4+offv2iomJ0f33368OHTpozJgxTb4uAK5DygEAXueuu+6Sv7+/kpOT1bFjR+Xn57vkuK+88opuuOEG3XnnnerVq5d+/etf64svvlCXLl0knZ5Fuu2229S7d2+NHDlSvXr10rx58yRJCQkJevjhh3XvvfcqJiZGf/zjHyVJjz76qB566CHNmTNHvXv31ogRI/Tee+8pKSnJ7tyPP/647rjjDqWlpamgoEDLli2zzV4B8AyGaZqmuwcBAK3VqlWrdNFFF+no0aNq27atu4cDoA7MNAEAADiBoAkAAMAJPJ4DAABwAjNNAAAATiBoAgAAcAJBEwAAgBMImgAAAJxA0AQAAOAEgiYAAAAnEDQBAAA4gaAJAADACQRNAAAATvj/bO1J78nACYkAAAAASUVORK5CYII=",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -890,7 +1078,7 @@
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHFCAYAAADv8c1wAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABWfklEQVR4nO3deVzU1f7H8dewuwAqKosrmhviBqShqW3icq9peYvKreV6o2spcrtpq9lm3nu7vxa37JplllqZqWVeaXFL1ETEDXcUVJBQA1zYZr6/P0xuBOKwyAzD+/l4zKM485nvfGZK5u35njlfk2EYBiIiIiJSJidbNyAiIiJSEyg0iYiIiFhBoUlERETECgpNIiIiIlZQaBIRERGxgkKTiIiIiBUUmkRERESsoNAkIiIiYgWFJhERERErKDSJiMMwmUxW3datW1ep53nxxRcxmUzXrPviiy+4//77ueGGG6hTpw6tW7dm5MiRHDp0qFLPLyK2YdJlVETEUWzZsqXYzy+//DI//PAD33//fbHxoKAgvLy8Kvw8J06c4MSJE9x0001l1vXq1Qs/Pz+GDx9OmzZtSE1N5bXXXiM1NZUtW7bQuXPnCvcgItVPoUlEHNaDDz7I559/zvnz58usu3jxInXr1q3y58/IyKBp06bFxk6dOkXr1q0ZM2YM//nPf6r8OUXk+tHpORGpVW655RaCg4PZsGEDvXv3pm7dujz88MMALF26lIiICPz9/alTpw6dOnViypQpXLhwodgxrD099/vABBAQEEDz5s1JTU2tmhckItXGxdYNiIhUt7S0NEaNGsVTTz3Fa6+9hpPT5b8/Hjp0iCFDhhAdHU29evXYv38/M2bMYNu2bSVO8VXU0aNHOX78OMOHD6+S44lI9VFoEpFa5+zZs3z22Wfcdtttxcafe+65on83DIM+ffrQqVMn+vfvz65du+jatWulnrewsJBHHnmE+vXrM2nSpEodS0Sqn07PiUit07BhwxKBCS7PAj3wwAP4+fnh7OyMq6sr/fv3ByApKalSz2kYBo888ggbN25k4cKFtGjRolLHE5Hqp5kmEal1/P39S4ydP3+evn374uHhwSuvvEL79u2pW7cuqamp3H333Vy6dKnCz2cYBn/+859ZtGgRH374IcOGDatM+yJiIwpNIlLrlLaI+/vvv+fUqVOsW7euaHYJ4JdffqnUc10JTAsWLGD+/PmMGjWqUscTEdvR6TkREf4XpNzd3YuNv/vuuxU+pmEYjBs3jgULFvDuu+/y0EMPVapHEbEtzTSJiAC9e/emYcOGREVFMXXqVFxdXfn4449JTEys8DEnTJjA/Pnzefjhh+nSpUuxzTfd3d3p0aNHVbQuItVEoUlEBPDx8eHrr7/mb3/7G6NGjaJevXoMGzaMpUuXEhISUqFjrlq1CoD333+f999/v9h9rVq14tixY5VtW0SqkXYEFxEpp0mTJvHRRx+RmZlp61ZEpBpppklExEoZGRnExcXxxRdfEB4ebut2RKSaaSG4iIiVVq9ezciRI2nXrh1vvfWWrdsRkWqm03MiIiIiVtBMk4iIiIgVFJpERERErKDQJCIiImIFfXuuilgsFk6dOoWnp2epl2gQERER+2MYBjk5OQQEBODkVPZckkJTFTl16pSuWi4iIlJDpaam0rx58zJrFJqqiKenJ3D5Tffy8rJxNyIiImKN7OxsWrRoUfQ5XhaFpipy5ZScl5eXQpOIiEgNY83SGi0EFxEREbGCQpOIiIiIFRSaRERERKyg0CQiIiJiBYUmERERESsoNImIiIhYQaFJRERExAoKTSIiIiJWUGgSERERsYJ2BBcRERG7ZrYYbEs+S0ZOLk09PegZ2Ahnp2vv4F3VFJpERETEbq3Zk8a0VftIy8otGvP39mDq0CAGBftXay86PSciIiJ2ac2eNB5btKNYYAJIz8rlsUU7WLMnrVr7UWgSERERu2O2GExbtQ+jlPuujE1btQ+zpbSK60OhSUREROzOtuSzJWaYfssA0rJy2ZZ8ttp6srvQNHv2bAIDA/Hw8CA0NJSNGzeWWb9+/XpCQ0Px8PCgTZs2zJ07t0TNsmXLCAoKwt3dnaCgIJYvX17s/hdffBGTyVTs5ufnV6WvS0RERKyXkXP1wFSRuqpgV6Fp6dKlREdH8+yzz5KQkEDfvn0ZPHgwKSkppdYnJyczZMgQ+vbtS0JCAs888wwTJkxg2bJlRTVxcXFERkYyevRoEhMTGT16NPfeey9bt24tdqzOnTuTlpZWdNu9e/d1fa0iIiJydU09Paq0riqYDMOovpOB19CrVy9CQkKYM2dO0VinTp0YPnw406dPL1E/efJkVq5cSVJSUtFYVFQUiYmJxMXFARAZGUl2djbffPNNUc2gQYNo2LAhixcvBi7PNH355Zfs3Lmzwr1nZ2fj7e1NVlYWXl5eFT6OiIiIwH82HuWVr5Ouer8J8PP2YNPk2yq1/UB5Pr/tZqYpPz+f+Ph4IiIiio1HRESwefPmUh8TFxdXon7gwIFs376dgoKCMmt+f8xDhw4REBBAYGAg9913H0ePHq3sSxIREZEK+ODH5GsGJoCpQ4Oqdb8muwlNmZmZmM1mfH19i437+vqSnp5e6mPS09NLrS8sLCQzM7PMmt8es1evXixcuJD//ve/vPfee6Snp9O7d2/OnDlz1X7z8vLIzs4udhMREZHK+XDzMV5ctQ+Av97Sljkje+DvXfwUnJ+3B3NGhVT7Pk12t7mlyVQ8MRqGUWLsWvW/H7/WMQcPHlz07126dCE8PJy2bdvy4YcfEhMTU+rzTp8+nWnTpl3j1YiIiIi1FsYdY+rKvQA8dktb/j6wAyaTiYjO/toR/LcaN26Ms7NziVmljIyMEjNFV/j5+ZVa7+Ligo+PT5k1VzsmQL169ejSpQuHDh26as3TTz9dLFBlZ2fTokWLq9aLiIjI1X0Ud4wXVlwOTFH92/LUr4EJwNnJRHhbH1u2B9jR6Tk3NzdCQ0OJjY0tNh4bG0vv3r1LfUx4eHiJ+rVr1xIWFoarq2uZNVc7Jlw+9ZaUlIS//9Wn/dzd3fHy8ip2ExERkfL7aMtxnv81MD3arw2TB3Uo8yyTrdhNaAKIiYnhP//5D++//z5JSUlMmjSJlJQUoqKigMuzO2PGjCmqj4qK4vjx48TExJCUlMT777/P/PnzefLJJ4tqJk6cyNq1a5kxYwb79+9nxowZfPvtt0RHRxfVPPnkk6xfv57k5GS2bt3Kn/70J7Kzsxk7dmy1vXYREZHa6OOtx3n+yz0A/KVfG6YM7miXgQns6PQcXN4e4MyZM7z00kukpaURHBzM6tWradWqFQBpaWnF9mwKDAxk9erVTJo0iVmzZhEQEMDbb7/NiBEjimp69+7NkiVLeO6553j++edp27YtS5cupVevXkU1J06c4P777yczM5MmTZpw0003sWXLlqLnFRERkar3ydYUnl1+OTCN6xvI03YcmMDO9mmqybRPk4iIiPUWb0vh6S8ubyT955sDefYPnWwSmGrkPk0iIiJSOyz5TWB6uI/tAlN5KTSJiIhItVn6UwpTfg1MD/VpzfN/rBmBCRSaREREpJp8+lNqUWB6sHdrXvhjUI0JTKDQJCIiItXg0+2pTP5iF4ZxOTBNHVqzAhMoNImIiMh19nn8CSYvuxyYxoa3qpGBCRSaRERE5Dr6PP4Ef/88EcOAMeGtePHOzjUyMIFCk4iIiFwnX+z4X2AadVNLptXgwAQKTSIiInIdLE84wd8+uxyYRvZqyUt3BtfowAQKTSIiIlLFvkw4yd8+vRyYHujVkpeHBePkVLMDEyg0iYiISBVasfMkMZ/uxGLA/T1b8oqDBCZQaBIREZEqsmLnSSYtvRyY7ruxBa8Od5zABApNIiIiUgVWJp4qFpheu6uLQwUmUGgSERGRSlqVeIroJQlYDIgMc8zABApNIiIiUglf7TpF9K8zTPeENmf63Y4ZmEChSURERCro611pTFyyE7PF4E+hzZkxoqvDBiZQaBIREZEKWL07jQlLEjBbDEaEOH5gAoUmERERKadvdqfxxOLLgenukGb8409dcXbwwAQKTSIiIlIOa/b8JjD1aMY//9StVgQmUGgSERERK63Zk87jnyRQaDG4q0cz/nlP7QlMoNAkIiIiVvjv3nQe/2QHhRaD4d0D+FctC0yg0CQiIiLXsHZvOuM/vhyYhnUP4I17u9e6wAQKTSIiIlKG2H2nGf/rDNPQbgG8UQtnmK5QaBIREZFSfbvvNH/9OJ4C8+XA9H/3dsPFufZGh9r7ykVEROSqvks6zWO/BqY/dvWv9YEJFJpERETkd77ff5rHFu2gwGzwhy7+vBnZvdYHJlBoEhERkd/4YX8GUR/tIN9sYUgXP968T4HpCr0LIiIiAsAPBzJ49KN48s0WBgf78dZ9PXBVYCqid0JERERY95vANKizH2/fr8D0e3o3REREarn1B3/mLx/Fk19oYWBnX955QIGpNC62bkBERESqj9lisC35LBk5uTT19CCv0FwUmCKCfHnn/hAFpqtQaBIREakl1uxJY9qqfaRl5Za4b0CQLzMfCMHNRYHpahSaREREaoE1e9J4bNEOjKvcP6ybvwLTNejdERERcXBmi8G0VfuuGpgAXl29H7OlrApRaBIREXFw25LPlnpK7rfSsnLZlny2mjqqmRSaREREHFxGTtmBqbx1tZVCk4iIiINr6ulRpXW1lRaCi4iIOLj0rEtl3m8C/Lw96BnYqHoaqqE00yQiIuLAPv0plZjPEq96v+nXf04dGoSzk+mqdaLQJCIi4rA+ijvGU8t2YRgwsldLZo/sgb938VNwft4ezBkVwqBgfxt1WXPo9JyIiIgD+s/Go7zydRIAD/cJ5Pk/dsJkMjGws3+xHcF7BjbSDJOVFJpEREQczDvfHeKN2IMAjL+1LU9GdMBkuhyMnJ1MhLf1sWV7NZZCk4iIiIMwDIN/rT3ArB+OAPC3Ae154vZ2Nu7KcSg0iYiIOADDMHjl6yTmb0oG4NkhnRjXr42Nu3IsCk0iIiI1nMVi8PyKPXy8NQWAl4Z1Zkx4a9s25YAUmkRERGows8Vg8rJdfB5/ApMJZtzdlXtvbGHrthySQpOIiEgNVWC2MGnpTr7alYazk4l/39uNYd2b2both6XQJCIiUgPlFZp54pME1u47jauzibfv68HgLtpr6XpSaBIREalhcgvMRC2KZ92Bn3FzcWLuqBBu6+hr67YcnkKTiIhIDXIxv5A/f7idzUfO4OHqxH/G3MjN7Rrbuq1aQaFJRESkhsjJLeChBT+x/fg56rk5s+ChnrrIbjVSaBIREakBfrmYz9j3t5F4IgsvDxc+fLgnPVo2tHVbtYpCk4iIiJ07cz6PUfO3kZSWTcO6rnz0SC+Cm3nbuq1aR6FJRETEjmVk5/LAf7ZyOOM8jeu788m4XrT39bR1W7WSQpOIiIidOvnLJUa+t4VjZy7i7+3Bx3/uRZsm9W3dVq2l0CQiImKHUs5c5P73tnDyl0s0b1iHxeNuokWjurZuq1ZTaBIREbEzhzPOM/I/WzidnUdg43p8/OdeBDSoY+u2aj0nWzfwe7NnzyYwMBAPDw9CQ0PZuHFjmfXr168nNDQUDw8P2rRpw9y5c0vULFu2jKCgINzd3QkKCmL58uVXPd706dMxmUxER0dX9qWIiIiU2/70bO6bF8fp7Dza+9Zn6aM3KTDZCbsKTUuXLiU6Oppnn32WhIQE+vbty+DBg0lJSSm1Pjk5mSFDhtC3b18SEhJ45plnmDBhAsuWLSuqiYuLIzIyktGjR5OYmMjo0aO599572bp1a4nj/fTTT8ybN4+uXbtet9coIiJyNXtOZnHfvC1kns8nyN+LJX8Jp6mnh63bkl+ZDMMwbN3EFb169SIkJIQ5c+YUjXXq1Inhw4czffr0EvWTJ09m5cqVJCUlFY1FRUWRmJhIXFwcAJGRkWRnZ/PNN98U1QwaNIiGDRuyePHiorHz588TEhLC7NmzeeWVV+jevTtvvvmm1b1nZ2fj7e1NVlYWXl5e5XnZIiIixB8/x4MLtpGTW0i3Fg1Y+FBPvOu62roth1eez2+7mWnKz88nPj6eiIiIYuMRERFs3ry51MfExcWVqB84cCDbt2+noKCgzJrfH3P8+PH84Q9/4I477rCq37y8PLKzs4vdREREKmLL0TOMmb+VnNxCbmzdkEWPKDDZI7sJTZmZmZjNZnx9i19w0NfXl/T09FIfk56eXmp9YWEhmZmZZdb89phLlixhx44dpc5mXc306dPx9vYuurVo0cLqx4qIiFyx4eDPPLhgGxfyzfS5wYcPH+6Jp4cCkz2ym9B0hclkKvazYRglxq5V//vxso6ZmprKxIkTWbRoER4e1p83fvrpp8nKyiq6paamWv1YERERgG/3nebPH24nt8DCbR2bMn/sjdR10xfb7ZXd/Jdp3Lgxzs7OJWaVMjIySswUXeHn51dqvYuLCz4+PmXWXDlmfHw8GRkZhIaGFt1vNpvZsGEDM2fOJC8vD2dn5xLP7e7ujru7e/lfqIiICPD1rjQmLkmg0GIwqLMfb9/fAzcXu5vLkN+wm/86bm5uhIaGEhsbW2w8NjaW3r17l/qY8PDwEvVr164lLCwMV1fXMmuuHPP2229n9+7d7Ny5s+gWFhbGyJEj2blzZ6mBSUREpDKWJ5zgicU7KLQYDOsewMwHFJhqAruZaQKIiYlh9OjRhIWFER4ezrx580hJSSEqKgq4fErs5MmTLFy4ELj8TbmZM2cSExPDuHHjiIuLY/78+cW+FTdx4kT69evHjBkzGDZsGCtWrODbb79l06ZNAHh6ehIcHFysj3r16uHj41NiXEREpLKWbEvh6eW7MQy4N6w50+/uirPT1ZehiP2wq9AUGRnJmTNneOmll0hLSyM4OJjVq1fTqlUrANLS0ort2RQYGMjq1auZNGkSs2bNIiAggLfffpsRI0YU1fTu3ZslS5bw3HPP8fzzz9O2bVuWLl1Kr169qv31iYhI7fbBj8m8uGofAKNvasW0OzvjpMBUY9jVPk01mfZpEhGRssxdf4TXv9kPwLi+gTwzpFOZX3SS6lGez2+7mmkSERFxNIZh8NZ3h3jz20MAPHHbDcQMaK/AVAMpNImIiFwnhmEwY80B5q4/AsDfB3Zg/K032LgrqSiFJhERkevAMAymrdrHB5uPAfD8H4N45OZA2zYllaLQJCIiUsUsFoNnv9zN4m2XNz5+ZXgwo25qZeOupLIUmkRERKpQodnCU5/v4ouEkziZYMaIrtwTpkttOQKFJhERkSpSYLYQvWQnX+9Ow9nJxP9FdufObgG2bkuqiEKTiIhIBZgtBtuSz5KRk0tTTw+6tfBmwuKdfJt0GldnEzMfCGFgZz9btylVSKFJRESknNbsSWPaqn2kZeUWjbm5OJFfaMHdxYm5o0O5tUNTG3Yo14NCk4iISDms2ZPGY4t28PudofMLLQA8dktbBSYHpasDioiIWMlsubyNQFmX0lj6Uypmiy624YgUmkRERKy0LflssVNypUnLymVb8tlq6kiqk0KTiIiIlTJyyg5M5a2TmkWhSURExEpNPT2qtE5qFi0EFxERsdL+9Owy7zcBft4e9AxsVD0NSbVSaBIREbkGs8Xg1a+TeP/H5KvWmH7959ShQTg7ma5aJzWXTs+JiIiU4VK+mccWxRcFpr8P7MCckT3w9y5+Cs7P24M5o0IYFOxvizalGmimSURE5CoycnIZ9+F2Ek9k4ebixBv3dGPor5dFiejsX2xH8J6BjTTD5OAUmkREREpx6HQODy74iZO/XKJhXVfeGxNGWOv/rVVydjIR3tbHhh1KdVNoEhER+Z3NhzN5dFE8ObmFtPapy4KHehLYuJ6t2xIbU2gSERH5jc/jTzBl2S4KLQZhrRoyb0wYjeq52botsQMKTSIiIoBhGPxf7EHe/v4wAEO7BfDPP3XFw9XZxp2JvVBoEhGRWi+v0Mzkz3fx5c5TAIy/tS1/G9ABJy3slt9QaBIRkVrtl4v5/OWjeLYln8XZycRrdwUTeWNLW7cldkihSUREaq3jZy7w0Ac/cfTnC3i6uzB7VAh92zWxdVtipxSaRESkVoo/fo5xC7dz9kI+Ad4eLHioJx38PG3dltgxhSYREal1vt6VRsynO8krtBDczIv3x95IUy9dZFfKptAkIiK1hmEYvLvhKK9/sx+AOzo15a37elDPXR+Hcm36v0RERGqFQrOFF1bu5ZOtKQA82Ls1z/9RF9cV6yk0iYiIwzufV8j4j3ew/uDPmEzw/B+CePjmQFu3JTWMQpOIiDi0tKxLPLTgJ/an5+Dh6sTb9/UgorOfrduSGkihSUREHNbeU1k8/MFPnM7Oo3F9d+aPDaNbiwa2bktqKIUmERFxSD/sz+DxT3ZwId9Mu6b1ef/BG2nRqK6t25IaTKFJREQczkdbjjN1xR4sBvS5wYfZI0PxruNq67akhlNoEhERh2GxGLy+Zj/zNhwF4J7Q5rx6VxfcXJxs3Jk4AoUmERFxCLkFZiYt3ck3e9IBeDKiPeNvvQGTSVsKSNVQaBIRkRov83wef/5wOztTf8HN2Yl/3tOVYd2b2botcTAKTSIiUqMdzjjPQx9sI/XsJRrUdeXdUaH0auNj67bEASk0iYhIjbXl6Bn+snA72bmFtPKpy4IHb6RNk/q2bksclEKTiIjUSF/sOMHkZbsoMBuEtGzAe2PC8Knvbuu2xIGVKzStXLmy3E8wYMAA6tSpU+7HiYiIlMYwDN767hBvfnsIgD908eeNe7vh4eps487E0ZUrNA0fPrxcBzeZTBw6dIg2bdqU63EiIiKlyS+0MOWLXXyx4yQAUf3b8tTADjjportSDcp9ei49PZ2mTZtaVevp6VnuhkREREqTdbGAqEXxxB09g7OTiVeGB3N/z5a2bktqkXKFprFjx5brVNuoUaPw8vIqd1MiIiK/lXr2Ig8u2MaRny9Q392FWSND6N++ia3bklrGZBiGYesmHEF2djbe3t5kZWUpKIqIVKGElHOMW7idzPP5+Ht78P6DN9LJX79npWqU5/Nb354TERG7tWZPGhOX7CSv0EKQvxfvP3gjft4etm5LaqlyhabAwMAKbUcfHR3NhAkTyv04ERGpHcwWg23JZ8nIyaWppwc3tm7IB5uP8erqJAwDbu3QhHceCKG+u/6uL7ZTrv/7Pvjggwo9SevWrSv0OBERcXxr9qQxbdU+0rJyi8bqujlzMd8MwOibWjF1aBAuzrrorthWuUJT//79r1cfIiJSC63Zk8Zji3bw+8W1VwLTn0Ka8dKwzrrortiFKovteXl5VXUoERGpBcwWg2mr9pUITL/145EzWPR1JbETVRaaevfuXWLs4MGDVXV4ERFxMNuSzxY7JVeatKxctiWfraaORMpW6RV1X331Ffv37+fChQucOnWKgICAovvuueceEhMTK/sUIiLigDJyyg5M5a0Tud4qHZo6d+5MSkoKGRkZ3H///aSmptK8eXMCAgJwdtZ1gEREpHSN6rlZVdfUU1sMiH2odGgKDAzkr3/9K8HBwfTr1w+AkydPkpycTHBwcKUbFBERx3PmfB4zvz9cZo0J8PP2oGdgo+ppSuQaKhSaCgoKGDRoEPXq1WPFihWYTKaiwATQrFkzmjVrVmVNioiI49hzMotHP4rn5C+X8HB1IrfAggmKLQi/8l25qUODcNbFeMVOVGghuKurKzt27CAlJUVfAxUREaut2HmSP83dzMlfLhHYuB5fPXEzc0eFlNjl28/bgzmjQhgU7G+jTkVKqvC35+666y4OHz7MiRMnqrIfZs+eTWBgIB4eHoSGhrJx48Yy69evX09oaCgeHh60adOGuXPnlqhZtmwZQUFBuLu7ExQUxPLly4vdP2fOHLp27YqXlxdeXl6Eh4fzzTffVOnrEhGpzcwWg+mrk5i4ZCe5BRZu6dCEL8f34YamngwK9mfT5NtYPO4m3rqvO4vH3cSmybcpMIndqfCapsaNG1NYWEhISAgjRozAz8+v2KzTCy+8UO5jLl26lOjoaGbPnk2fPn149913GTx4MPv27aNly5Yl6pOTkxkyZAjjxo1j0aJF/Pjjj/z1r3+lSZMmjBgxAoC4uDgiIyN5+eWXueuuu1i+fDn33nsvmzZtolevXgA0b96c119/nRtuuAGADz/8kGHDhpGQkEDnzp0r8vaIiMivsi4W8PjiHWw8lAnAY7e05cmIDsVOuzk7mQhv62OrFkWsYjIMo0Lbhjk5OWEymTAMo9RTdGazudzH7NWrFyEhIcyZM6dorFOnTgwfPpzp06eXqJ88eTIrV64kKSmpaCwqKorExETi4uIAiIyMJDs7u9jM0aBBg2jYsCGLFy++ai+NGjXin//8J4888ohVvZfnKskiIrXFwdM5jFu4neNnLlLH1Zl/3tOVP3YNuPYDRapJeT6/KzzT1LJlyypdz5Sfn098fDxTpkwpNh4REcHmzZtLfUxcXBwRERHFxgYOHMj8+fMpKCjA1dWVuLg4Jk2aVKLmzTffLPWYZrOZzz77jAsXLhAeHn7VfvPy8ortgp6dnV3WyxMRqXX+uzedmKU7uZBvplmDOrw3JoygAP2lUmquCoemY8eOVWEbkJmZidlsxtfXt9i4r68v6enppT4mPT291PrCwkIyMzPx9/e/as3vj7l7927Cw8PJzc2lfv36LF++nKCgoKv2O336dKZNm1aelygiUitYLAZvfXeIt747BEB4Gx9mjQyxel8mEXtld5eM/v3s1dVO/5VV//txa47ZoUMHdu7cyZYtW3jssccYO3Ys+/btu+rzPv3002RlZRXdUlNTy35hIiK1QE5uAY8uii8KTA/1ac3CR3oqMIlDqPTmllWlcePGODs7l5gBysjIKDFTdIWfn1+p9S4uLvj4+JRZ8/tjurm5FS0EDwsL46effuKtt97i3XffLfW53d3dcXd3t/4Fiog4uOTMC4xbuJ3DGedxc3Hi1eHB3BPWwtZtiVQZu5lpcnNzIzQ0lNjY2GLjsbGxpV4MGCA8PLxE/dq1awkLC8PV1bXMmqsd8wrDMIqtWRIRkatbdyCDO2du4nDGeXy93Pn00XAFJnE4VTLTtGPHDoKDg3Fzq9z0a0xMDKNHjyYsLIzw8HDmzZtHSkoKUVFRwOVTYidPnmThwoXA5W/KzZw5k5iYGMaNG0dcXBzz588v9q24iRMn0q9fP2bMmMGwYcNYsWIF3377LZs2bSqqeeaZZxg8eDAtWrQgJyeHJUuWsG7dOtasWVOp1yMi4ugMw2Du+qP847/7MQwIadmAuaNCaeql68WJ46mS0HTjjTeSlJRE+/btK3WcyMhIzpw5w0svvURaWhrBwcGsXr2aVq1aAZCWlkZKSkpRfWBgIKtXr2bSpEnMmjWLgIAA3n777aI9mgB69+7NkiVLeO6553j++edp27YtS5cuLdqjCeD06dOMHj2atLQ0vL296dq1K2vWrGHAgAGVej0iIo7sUr6Zp5btYlXiKQDu79mCF+/sjLuLLtYujqnC+zT9lpOTE/v37690aKrJtE+TiNQmqWcv8uhH8exLy8bFycTUOzszqlfVbkUjUh2qZZ8mERGpneKOnGH8Jzs4eyEfn3puzB4ZQq822s1bHJ9Ck4iIWMUwDD7cfIyXv07CbDEIbubFvNFhBDSoY+vWRKqFQpOIiFxTboGZ57/cw2fxly/SPrx7AK+P6IqHq9YvSe2h0CQiImU6nZ3Lox/FszP1F5xM8PTgTvy5b6DWL0mto9AkIiJXFX/8HFGL4vk5Jw/vOq7MfKAHfds1sXVbIjah0CQiIqVasi2F51fsocBs0MHXk3ljQmnlU8/WbYnYTJWEpqlTp9K4ceOqOJSIiNhYfqGFl7/ax0dbjgMwONiPf93TjXru+nu21G5Vsk+TaJ8mEXEMmefz+OuiHWw7dhaTCf42oD3jb71B65fEYWmfJhERKbfdJ7J49KPtnMrKpb67C29GdueOoNIvmC5SGyk0iYgIXyacZPKyXeQVWmjTuB7zxoRxQ9P6tm5LxK4oNImI1GKFZgsz1uznvY3JANzaoQlv3tcD7zquNu5MxP5USWj65Zdf+O9//8vJkycxmUz4+/szcOBAGjZsWBWHFxGR6+CXi/k8sTiBjYcyARh/a1tiBnTA2Unrl0RK41TZA8yfP5+ePXuyZcsWLBYLZrOZLVu2cNNNNzF//vyq6FFERKrYgfQc7pz5IxsPZVLH1ZlZD4Tw94EdFZhEylDpb8916NCB+Ph46tcvfu47JyeH0NBQDh48WKkGawp9e05Eaoo1e9KI+TSRi/lmmjesw7zRYQQF6PeW1E7V+u05k8nE+fPnS4Sm8+fP6yuqIiJ2xGIxePPbg7z9/WEAerf1YdYDITSs52bjzkRqhkqHpn/961/079+f4OBgmjVrBsCJEyfYu3cvb7zxRqUbFBGR8jFbDLYlnyUjJ5emnh70DGzExfxCJi3dybdJGQA83CeQZ4Z0xMW50qs0RGqNKtnc0mw2s23bNk6dOoVhGDRr1oyePXvi7Fx7rn6t03MiYg/W7Elj2qp9pGXlFo01qe+Gs5OJ9Ow83FycmH5XF0aENrdhlyL247qfnisoKGDQoEHUq1ePFStW4OzsTHh4eIWaFRGRqrFmTxqPLdrB7/8m/PP5fAAa1HXlw4d60q1Fg2rvTcQRVGhe1tXVlR07dpCSkqJ1SyIidsBsMZi2al+JwPRbbs5OBDfzrraeRBxNhU9m33XXXRw+fJgTJ05UZT8iIlIB25LPFjslV5qMnDy2JZ+tpo5EHE+FF4I3btyYwsJCQkJCGDFiBH5+fsVmnV544YUqaVBERK4tI6fswFTeOhEpqcILwZ2cnDCZTBiGUeopOrPZXOnmahItBBcRW4o7cob739tyzbrF424ivK1PNXQkUjNUyz5NLVu21HomERE7YLYY/Hg4s8waE+DnfXn7ARGpmAqHpmPHjlVhGyIiUhG/XMwneulO1h34+ao1V/56O3VokC6TIlIJVXLBXhERqX77TmXz6KLtpJ69hIerE9Pv7kIdV+cS+zT5eXswdWgQg4L9bditSM2n0CQiUgN9mXCSKV/sIrfAQvOGdXh3dCidAy5vJzAgyK/EjuCaYRKpvHKFpsDAwAqtY4qOjmbChAnlfpyIiBRXYLbw6tdJfLD5GAD92jfh7fu606Du/64f5+xk0mJvkeugXKHpgw8+qNCTtG7dukKPExGR/8nIyeXxjxPYduzyXktP3HYD0Xe01yySSDUpV2jq37//9epDRETKEH/8HH/9OJ7T2XnUd3fh3/d2I6Kzn63bEqlVyrUj+K5du7BYLFbX7927l8LCwnI3JSIilxmGwUdbjnPfvDhOZ+fRrml9VjzeR4FJxAbKFZp69OjBmTNnrK4PDw8nJSWl3E2JiAjkFpj5++e7eP7LPRSYDYZ08WP5+D60bVLf1q2J1ErlOj1nGAbPP/88devWtao+Pz+/Qk2JiNR2J85dJGpRPHtOZuNkgsmDOvKXfm20qbCIDZUrNPXr148DBw5YXR8eHk6dOnXK3ZSISG228dDPTFicwLmLBTSs68rMB0Loc0NjW7clUuuVKzStW7fuOrUhIiKGYTBn/RH+9d8DWAzo0sybOaNCaN7Qutl9Ebm+tLmliIgdOJ9XyJOfJrJmbzoA94Q25+XhwXi4Otu4MxG5osKhyWKxsGDBAr777jtOnz6NYRhF95lMJr777rsqaVBExNEdzjjPox9t58jPF3B1NvHinZ15oKcuii5ibyocmmJiYnjnnXcAigKTyWTCMAz9QRcRsdKaPek8+Vki5/MK8fVyZ86oUEJaNrR1WyJSigqHpsWLF2MYBgEBAQQGBuLiojN9IiLWMlsM3lh7gNnrjgDQM7ARsx4IoYmnu407E5GrqXDSMZvNNG/enEOHDuHurj/kIiLWOnchnwlLEth4KBOAR24OZMrgjrg6l2vrPBGpZhX+E3rfffdx6dIlCgoKqrIfERGHtudkFkNnbmLjoUw8XJ14677uPP/HIAUmkRqgwjNN9evXJzs7m+7du3PnnXfSoEGDYve/8MILle1NRMShLIs/wTPLd5NXaKGVT13mjgqlk7+XrdsSESuZjN9+7a0cnJycylz4bTabK91cTZKdnY23tzdZWVl4eemXoIj8T36hhVe+3sfCuOMA3NqhCW9G9sC7rquNOxOR8nx+V3imqWVLfR1WRORaTmfn8tePdxB//BwAE29vx8Tb2+HkpN+fIjVNhUPTsWPHqrANERHH89Oxs/z14x38nJOHp4cLb0Z25/ZOvrZuS0QqSPsEiIhUMcMw+HDzMV75OolCi0EHX0/mjg4lsHE9W7cmIpVQqdCUn5/Pjz/+yKlTp0qsYRozZkylGhMRqYku5Zt5dvluvkg4CcAfu/ozY0RX6rnr76giNV2F/xQfOnSIAQMGkJqaWuI+k8mk0CQitU7KmYs8uiiepLRsnJ1MPD24I4/cHKj1nyIOosKhacqUKaSkpFRlLyIiNda6AxlMXLKTrEsF+NRzY+YDIYS39bF1WyJShSq8m9qmTZtwcXEhNjYWgB49erB48WIaN25cNCYi4ugsFoOZ3x/ioQ9+IutSAd1aNOCrCTcrMIk4oAqHpnPnztGpUyduv/12TCYTrq6uREZG4ufnx2uvvVaVPYqI2KXs3AIeXRTPv9YexDDg/p4t+fTRm/D3rmPr1kTkOqjw6TlPT08sFgtweXfw/fv3s3XrVlJSUjhy5EiVNSgiYo8Onc7h0Y/iOZp5ATdnJ14a1pn7era0dVsich1VanPLI0eOYDab6dKlC3FxcfTu3RuAwMDAKmtQRMSWzBaDbclnycjJpamnBz0DG/Hfvek8+VkiF/PN+Ht7MGdUKN1bNLB1qyJynVU4NI0cOZL169dz8OBBnn32We666y7y8/NxdnbmxRdfrMIWRURsY82eNKat2kdaVm7RWD03Zy7kX95iJbyND+880IPG9d1t1aKIVKMKX3vu95KTk0lISKBz58506NChKg5Zo+jacyKOZc2eNB5btIOr/YIc0Kkpc0aF4uJc4aWhImIHquXacwCZmZm88847bNmyhVatWjFhwgS2bt1KnTp1aNlS5/ZFpGYyWwymrdp31cAEsOdUtvZfEqllKvxXpGPHjtGtWzdeeeUVvv32W/bs2UNWVhYPPvggM2fOrHBDs2fPJjAwEA8PD0JDQ9m4cWOZ9evXryc0NBQPDw/atGnD3LlzS9QsW7aMoKAg3N3dCQoKYvny5cXunz59OjfeeCOenp40bdqU4cOHc+DAgQq/BhGp2bYlny12Sq40aVm5bEs+W00diYg9qHBoeuqpp0hLS6NZs2ZcOcPXp08fvLy8KrxP09KlS4mOjubZZ58lISGBvn37Mnjw4KtuopmcnMyQIUPo27cvCQkJPPPMM0yYMIFly5YV1cTFxREZGcno0aNJTExk9OjR3HvvvWzdurWoZv369YwfP54tW7YQGxtLYWEhERERXLhwoUKvQ0RqtoycsgNTeetExDFUeE1To0aNcHFxITk5GU9PT2666SY2b95Mt27dSElJ4dy5c+U+Zq9evQgJCWHOnDlFY506dWL48OFMnz69RP3kyZNZuXIlSUlJRWNRUVEkJiYSFxcHQGRkJNnZ2XzzzTdFNYMGDaJhw4YsXry41D5+/vlnmjZtyvr16+nXr59VvWtNk4jjiN17mnEfbb9m3eJxN2kTS5Earjyf3xWeabp06RJ+fn7Uq1f8qt3nz58nLy+v3MfLz88nPj6eiIiIYuMRERFs3ry51MfExcWVqB84cCDbt2+noKCgzJqrHRMgKysLuBwMryYvL4/s7OxiNxGp+fanZ/Py1/vKrDEB/t6Xtx8QkdqjwqGpbdu27N27l0WLFgGXQ8Q777xDcnIy7du3L/fxMjMzMZvN+Pr6Fhv39fUlPT291Mekp6eXWl9YWEhmZmaZNVc7pmEYxMTEcPPNNxMcHHzVfqdPn463t3fRrUWLFtd8jSJi39bsSePu2ZtJOXuRxvXdgMsB6beu/Dx1aBDOTloILlKbVDg0jRs3DsMwGDt2LCaTiZ07dxIdHY3JZOLhhx+ucEO//zaKYRhlfkOltPrfj5fnmI8//ji7du266qm7K55++mmysrKKbqmpqWXWi4j9slgM/r32AFGLdnAx30zvtj7ETurP3FEh+Hl7FKv18/ZgzqgQBgX726hbEbGVCm85MGHCBA4cOMC7775bLKiMGzeOCRMmlPt4jRs3xtnZucQMUEZGRomZoiv8/PxKrXdxccHHx6fMmtKO+cQTT7By5Uo2bNhA8+bNy+zX3d0dd3dtaCdS0+XkFjBpaSLfJp0G4OE+gTwzpCMuzk4MCvZnQJBfiR3BNcMkUjtVODSZTCZmz57NU089xfbt2zEMg7CwsApfQsXNzY3Q0FBiY2O56667isZjY2MZNmxYqY8JDw9n1apVxcbWrl1LWFgYrq6uRTWxsbFMmjSpWM2VS77A5ZmnJ554guXLl7Nu3TpdBkakljj683n+8lE8hzPO4+bixPS7ujAitPhfmJydTFrsLSJAJTe3NJvN5Ofn4+vri2EYpKamFp2msvZbZ78VExPD6NGjCQsLIzw8nHnz5pGSkkJUVBRw+ZTYyZMnWbhwIXD5m3IzZ84kJiaGcePGERcXx/z584udWps4cSL9+vVjxowZDBs2jBUrVvDtt9+yadOmoprx48fzySefsGLFCjw9PYtmpry9valTR1crF3FEPxzIYMLiBHJyC/Hz8mDuaF0/TkSuwaigH3/80WjVqpXh5ORU4ubs7FzRwxqzZs0yWrVqZbi5uRkhISHG+vXri+4bO3as0b9//2L169atM3r06GG4ubkZrVu3NubMmVPimJ999pnRoUMHw9XV1ejYsaOxbNmyYvcDpd4WLFhgdd9ZWVkGYGRlZZXr9YpI9bJYLMbsHw4brad8ZbSa/JVx9+wfjdPZl2zdlojYSHk+vyu8T1P37t3ZtWtXqfeZTCbMZnOFQlxNpX2aROzfpXwzTy3bxarEUwDc37MFL97ZGXcXZxt3JiK2Ui3Xnjt8+DANGzbks88+o02bNroGk4jYtRPnLvKXhfHsS8vGxcnE1Ds7M6pXS/3uEhGrVTg09e/fn927d9OvXz9cXCq1NEpE5LqKO3KG8Z/s4OyFfHzquTF7ZAi92mhxt4iUT4VPz6Wnp3PLLbfg5uZGREREiSmtF154oUoarCl0ek7E/hiGwcK447z01T7MFoPgZl68OzqMZg30BQ8RuaxaTs+tWbOGI0eOYLFY2Lt3b4n7a1toEhH7kldo5vkv9/Dp9hMADOsewOt3d6WOm9YviUjFVDg0Pffcc7VusbeI1AwZ2bk8uiiehJRfcDLBlMEdGddXay9FpHIqfBmV8+fP4+/vz6FDhygoKMBisRS7iYjYQkLKOf74ziYSUn7By8OFBQ/15C/92iowiUilVTg0PfLIIxQWFtK0aVOcnTXdLSK29+n2VCLf3UJGTh7tmtZn5eM30799E1u3JSIOosKn537++Weys7Np164dN998c7HFUyaTifnz51dJgyIi11JgtvDq10l8sPkYABFBvvw7sjv13fXNXhGpOhX+9pyTkxMmkwnDMIpNe1/5ubatd9K350Rs4+yFfMZ/vIO4o2cAiL6jHRNua4eTLqorIlaolm/PtWypTeFExLb2ncrmLx9t58S5S9Rzc+bfkd0Z2NnP1m2JiIOqcGg6duxYFbYhIlI+qxJP8ffPE8ktsNDKpy7vjQmjva+nrdsSEQemE/4iUqOYLQZvrD3A7HVHAOjbrjEz7w/Bu66rjTsTEUen0CQiNUbWpQKilyTww4GfAXi0XxueGtQRZ61fEpFqoNAkIjXC4Yzz/GXhdo5mXsDdxYkZI7oyvEczW7clIrWIQpOI2L3vkk4TvWQnOXmFBHh78O7oMLo097Z1WyJSyyg0iYjdMgyDWT8c5o3YgxgG9GzdiNmjQmhc393WrYlILaTQJCJ26UJeIX//PJHVu9MBGHVTS174Y2fcXCp8IQMRkUpRaBIRu5N69iLjFm5nf3oOrs4mXhoWzP09W9q6LRGp5RSaRMSu/Hg4k/Gf7OCXiwU0ru/O3FEhhLVuZOu2REQUmkTEPhiGwfs/HuO11UmYLQbdmnszd3Qo/t51bN2aiAig0CQidiC3wMyzy/ewbMcJAO4OacZrd3XBw9XZxp2JiPyPQpOIVBuzxWBb8lkycnJp6ulBz8BG/JyTx6MfbSfxRBbOTiaeGdKJh/u01rUtRcTuKDSJSLVYsyeNaav2kZaVWzTmU8+NArOF7NxCGtR1ZdYDIfS5obENuxQRuTqFJhG57tbsSeOxRTswfjd+5kI+AM0aeLB4XDgtfepWf3MiIlbShicicl2ZLQbTVu0rEZh+q9Bi0KyhFnyLiH1TaBKR62pb8tlip+RKczo7j23JZ6upIxGRilFoEpHrKiOn7MBU3joREVtRaBKR66qpp0eV1omI2IoWgovIdVNotvBt0ukya0yAn/fl7QdEROyZQpOIXBfnLuTzxOIENh3OvGrNlZ2Ypg4NwtlJ+zKJiH1TaBKRKpeUls1fPtpO6tlL1HVz5l/3dMPJRIl9mvy8PZg6NIhBwf427FZExDoKTSJSpb7elcaTnyVyqcBMy0Z1mTcmlI5+XgAMCPIrsSO4ZphEpKZQaBKRKmG2GPw79gCzfjgCQN92jXnn/h40qOtWVOPsZCK8rY+tWhQRqRSFJhGptKxLBUQvSeCHAz8D8Jd+bXhqYAdcnPUFXRFxHApNIlIphzNyGLcwnuTMC7i7ODFjRFeG92hm67ZERKqcQpOIVFjsvtNMWrqT83mFBHh7MG9MGMHNvG3dlojIdaHQJCLlZrEYvPP9Yf7v24MA9ApsxKyRITSu727jzkRErh+FJhEpl/N5hfzt0538d+/lTSvHhrfiuT8G4ar1SyLi4BSaRMRqxzIvMG7hdg5lnMfN2YlXhgdz740tbN2WiEi1UGgSEausO5DBhMUJZOcW4uvlztxRofRo2dDWbYmIVBuFJhEpk2EYvLvhKP9Ysx+LASEtGzB3VChNvXSBXRGpXRSaROSqLuYX8tTnu/hqVxoA993YgmnDOuPu4mzjzkREqp9Ck4iUKvXsRf7yUTxJadm4OJl48c7OjOzVEpNJlz0RkdpJoUlESth8OJPxn+zg3MUCGtd3Y/bIUHoGNrJ1WyIiNqXQJCJFDMNgwY/HeHV1EmaLQZdm3rw7OpSABnVs3ZqIiM0pNIkIALkFZp5dvodlO04AcHePZrx2dxc8XLV+SUQEFJpEBEjLukTUR/EknsjC2cnEM0M68XCf1lq/JCLyGwpNIrXcT8fO8tiieDLP59OgriuzHgihzw2Nbd2WiIjdUWgSqcU+3nqcF1fupcBs0NHPk/fGhNGiUV1btyUiYpcUmkRqofxCC1NX7mXxthQA/tDVn3/+qSt13fQrQUTkavQbUqSWycjO5bGPdxB//BwmEzw1sCNR/dto/ZKIyDUoNInUIjtTf+HRj7ZzOjsPTw8X3r6/B7d2aGrrtkREagSFJpFa4rPtqTz75R7yCy3c0LQ+740JI7BxPVu3JSJSYyg0iTi4ArOFV79O4oPNxwAYEOTLv+/thqeHq20bExGpYRSaRBzYmfN5PP5JAnFHzwAQfUc7JtzWDicnrV8SESkvJ1s38HuzZ88mMDAQDw8PQkND2bhxY5n169evJzQ0FA8PD9q0acPcuXNL1CxbtoygoCDc3d0JCgpi+fLlxe7fsGEDQ4cOJSAgAJPJxJdfflmVL0nEJvaczOLOmT8Sd/QM9dycmTc6lOg72iswiYhUkF2FpqVLlxIdHc2zzz5LQkICffv2ZfDgwaSkpJRan5yczJAhQ+jbty8JCQk888wzTJgwgWXLlhXVxMXFERkZyejRo0lMTGT06NHce++9bN26tajmwoULdOvWjZkzZ1731yhSHVbsPMmf5m7m5C+XaO1Tly/H9yGis5+t2xIRqdFMhmEYtm7iil69ehESEsKcOXOKxjp16sTw4cOZPn16ifrJkyezcuVKkpKSisaioqJITEwkLi4OgMjISLKzs/nmm2+KagYNGkTDhg1ZvHhxiWOaTCaWL1/O8OHDy9V7dnY23t7eZGVl4eXlVa7HilSE2WKwLfksGTm5NPX0oGdgIwD+8d/9vLv+KAC3dGjCW/f1wLuO1i+JiJSmPJ/fdrOmKT8/n/j4eKZMmVJsPCIigs2bN5f6mLi4OCIiIoqNDRw4kPnz51NQUICrqytxcXFMmjSpRM2bb75Zpf2LVKc1e9KYtmofaVm5RWO+nu741HdjX1oOAI/d0pYnIzrgrNNxIiJVwm5CU2ZmJmazGV9f32Ljvr6+pKenl/qY9PT0UusLCwvJzMzE39//qjVXO6a18vLyyMvLK/o5Ozu7UscTsdaaPWk8tmgHv58iPp2Tx+mcPNycnfh3ZDf+2DXAJv2JiDgqu1rTBJTYldgwjDJ3Ki6t/vfj5T2mNaZPn463t3fRrUWLFpU6nog1zBaDaav2lQhMv+VVx4XBwf7V1pOISG1hN6GpcePGODs7l5gBysjIKDFTdIWfn1+p9S4uLvj4+JRZc7VjWuvpp58mKyur6Jaamlqp44lYY1vy2WKn5EqTeT6fbclnq6kjEZHaw25Ck5ubG6GhocTGxhYbj42NpXfv3qU+Jjw8vET92rVrCQsLw9XVtcyaqx3TWu7u7nh5eRW7iVxvGTllB6by1omIiPXsZk0TQExMDKNHjyYsLIzw8HDmzZtHSkoKUVFRwOXZnZMnT7Jw4ULg8jflZs6cSUxMDOPGjSMuLo758+cX+1bcxIkT6devHzNmzGDYsGGsWLGCb7/9lk2bNhXVnD9/nsOHDxf9nJyczM6dO2nUqBEtW7asplcvcm1NPT2qtE5ERKxnV6EpMjKSM2fO8NJLL5GWlkZwcDCrV6+mVatWAKSlpRXbsykwMJDVq1czadIkZs2aRUBAAG+//TYjRowoqunduzdLlizhueee4/nnn6dt27YsXbqUXr16FdVs376dW2+9tejnmJgYAMaOHcsHH3xwnV+1iHUu5BWybMeJMmtMgJ/3/7YfEBGRqmNX+zTVZNqnSa6nxNRfmLgkgWNnLmIyQWl/aq98tWHOqBAGaSG4iIhVauQ+TSJSksViMG/jUf713wMUWgwCvD34v8junLuYX2KfJj9vD6YODVJgEhG5ThSaROxUelYuMZ/uZPORyxfbHdLFj+l3dcW77uUvOQwI8iuxI7g2shQRuX4UmkTs0Nq96UxetotzFwuo4+rMtDs7c09Y82L7izk7mQhv62PDLkVEaheFJhE7cinfzCtf7+PjrZe/8BDczIu37utB2yb1bdyZiIgoNInYiX2nspmwJIHDGecBeLRfG/4W0QE3F7vZTk1EpFZTaBKxMcMwWPDjMV7/Zj/5ZgtNPd35973dubldY1u3JiIiv6HQJGJDmefzePKzRNYd+BmAOzo1ZcaIrvjUd7dxZyIi8nsKTSI2su5ABk9+lkjm+XzcXZx47g+dGHVTq0pfTFpERK4PhSaRapZXaGbGNwd4/8dkADr4evL2/T3o4Odp485ERKQsCk0i1ehwRg5PLN5JUlo2AA/2bs2UwR3xcHW2cWciInItCk0i1cAwDD7ZlsLLX+0jt8BCo3pu/OuertzW0dfWrYmIiJUUmkSus3MX8pnyxS7+u/c0AH3bNeaNe7rR1MvDxp2JiEh5KDSJXEebD2cS82ki6dm5uDqbmDyoIw/3CcRJlzsREalxFJpEroMCs4V/xx5k7vojGAa0aVKPt+/rQXAzb1u3JiIiFaTQJFLFjmVeYOKSBBJPZAFwf88WPP/HIOq66Y+biEhNpt/iIlXEMAyW7TjJ1BV7uJBvxruOK6/f3YXBXfxt3ZqIiFQBhSaRKpB1qYDnvtzDqsRTAPQKbMT/RXYnoEEdG3cmIiJVRaFJpJK2HzvLxCU7OfnLJZydTMQMaE9U/7Y4a7G3iIhDUWgSqaBCs4WZPxzm7e8OYTGgZaO6vHVfd3q0bGjr1kRE5DpQaBKpgBPnLhK9ZCfbj58D4O4ezZg2rDOeHq427kxERK4XhSaRclqVeIpnlu8mJ7eQ+u4uvDI8mOE9mtm6LRERuc4UmkSsdD6vkBdX7uXz+BMA9GjZgLcie9DSp66NOxMRkeqg0CRihcTUX5i4JIFjZy7iZILHb72BCbe3w8XZydatiYhINVFoEimDxWLw7oajvLH2AIUWgwBvD/4vsju92vjYujUREalmCk0igNlisC35LBk5uTT19KBnYCN+zskj5tOdbD5yBoAhXfyYfldXvOtqsbeISG2k0CS13po9aUxbtY+0rNyisYZ1XckvtHAh30wdV2em3dmZe8KaYzJp7yURkdpKoUlqtTV70nhs0Q6M342fu1gAQMtGdVjwUE/aNqlf/c2JiIhd0SpWqbXMFoNpq/aVCEy/lV9oobVPvWrrSURE7JdCk9Ra25LPFjslV5r07Dy2JZ+tpo5ERMSeKTRJrbU/PduquoycsoOViIjUDlrTJLXOqV8uMeuHwyz5KdWq+qaeHte5IxERqQkUmqTWSM/KZfa6wyzZlkq+2QKAm7NT0b//ngnw8768/YCIiIhCkzi8jOxcZq87wifbUsgvvByQegY2ImZAe365mM9ji3YAFFsQfmVjgalDg3B20jYDIiKi0CQO7OecPOauP8KiLcfJ+zUs3di6IZPuaE94W5+iPZfmjAopsU+Tn7cHU4cGMSjY3ya9i4iI/VFoEoeTeT6PeRuOsjDuGLkFl8NSSMsGxAzoQJ8bfEpsUDko2J8BQX4ldgTXDJOIiPyWQpM4jLMX8pm34Sgfbj7GpQIzAN1aNCBmQHv6tWtc5m7ezk4mwtvqenIiInJ1Ck1S4/1yMZ/3Nh7lgx+PcSH/cljq0sybmAHtuaVDE136REREqoRCk9RYWRcLmL/pKO//eIzzeYUAdA7wYtId7bm9U1OFJRERqVIKTVLjZOcW8P6mZOZvSiYn93JY6uTvRfQd7YgI8lVYEhGR60KhSWqMnNwCFvx4jP9sPEr2r2Gpg68n0Xe0Y2BnP5y0cFtERK4jhSaxe+fzCvlw8zHe23iUXy4WANCuaX0m3tGOIcH+CksiIlItFJrEbl3IK2Rh3HHmbTjCuV/DUpsm9Yi+oz1/6OKvLQFERKRaKTSJ3bmUb+ajLcd4d/1RzlzIByCwcT0m3t6Ood0CFJZERMQmFJrEbuQWmFm05Thz1x8l83weAK186jLhtnYM6x6Ai7OTjTsUEZHaTKFJbC63wMzibSnMWXeEjJzLYalFozo8cVs77urRDFeFJRERsQMKTWIzeYVmlv6UyuwfjpCeffm6b80a1OGJ225gRGhzhSUREbErCk1S7fILLXy6PZVZPxwuukhugLcH42+7gXtCW+DmorAkIiL2R6FJqpzZYpR68dsCs4XP408w8/vDnPzlEgB+Xh6Mv7Ut997YAncXZxt3LiIicnUKTVKl1uxJY9qqfUUzSAB+Xu4MCPJl3cGfST17OSw19XTnr7e05b6eLfFwVVgSERH7p9AkVWbNnjQeW7QD43fj6dl5fLQlBYDG9d157Ja2jOylsCQiIjWLQpNUCbPFYNqqfSUC0295eriw7slbqO+h/+1ERKTm0aeXVFjWxQKS0rNJSstm/YGfi52SK01ObiG7T2YR3tanmjoUERGpOgpNck0Wi8GxMxdISsshKS2b/enZJKXlFC3mLo+MnLKDlYiIiL1SaJJicnILOJB+ORzt+zUkHUjP4VKBudT6Zg3q0MnfCy8PF75IOHnN4zf19KjqlkVERKqFQlMtZbEYnDh3iX1pl0+vJaVlk5SeXfTttt/zcHWig68nnfy9im4d/T3x8nAFLq9pijt6hvSs3FLXNZkAP+/L2w+IiIjURApNdu5qex6Vx8X8Qvan57D/15mjy6fYcjifV1hqvb+3x6/ByJOOfpcDUmDjemU+r7OTialDg3hs0Q5MUCw4XXnU1KFButiuiIjUWHYXmmbPns0///lP0tLS6Ny5M2+++SZ9+/a9av369euJiYlh7969BAQE8NRTTxEVFVWsZtmyZTz//PMcOXKEtm3b8uqrr3LXXXdV6nmrQ2l7Hvl7ezB1aBCDgv1L1BuGwamsXJJO/W/mKCkth2NnLmCUMv3j5uxEO9/6v5k98qSTnxcN67lVqN9Bwf7MGRVScp+mMnoWERGpKewqNC1dupTo6Ghmz55Nnz59ePfddxk8eDD79u2jZcuWJeqTk5MZMmQI48aNY9GiRfz444/89a9/pUmTJowYMQKAuLg4IiMjefnll7nrrrtYvnw59957L5s2baJXr14Vet7qcNU9j7JyeWzRDt66rzutG9f7debofzNI2bmlzx418XQvCkZB/v+bParq67sNCvZnQJBfpWfHRERE7I3JMEqbg7CNXr16ERISwpw5c4rGOnXqxPDhw5k+fXqJ+smTJ7Ny5UqSkpKKxqKiokhMTCQuLg6AyMhIsrOz+eabb4pqBg0aRMOGDVm8eHGFnrc02dnZeHt7k5WVhZeXV/le+O+YLQY3z/j+ml/hL42Lk4kbmtYvCkYd/S+vQ2pc371SPYmIiDii8nx+281MU35+PvHx8UyZMqXYeEREBJs3by71MXFxcURERBQbGzhwIPPnz6egoABXV1fi4uKYNGlSiZo333yzws8LkJeXR15eXtHP2dnZ13yN1tqWfNaqwOTp4ULX5t508vvf4uwbmtbXBW9FRESuA7sJTZmZmZjNZnx9fYuN+/r6kp6eXupj0tPTS60vLCwkMzMTf3//q9ZcOWZFnhdg+vTpTJs2zerXVx7W7mX08rDODO/R/Lr0ICIiIsXZ3ZSEyVR87YthGCXGrlX/+3Frjlne53366afJysoquqWmpl61trys3cvI16tOlT2niIiIlM1uZpoaN26Ms7NzidmdjIyMErNAV/j5+ZVa7+Ligo+PT5k1V45ZkecFcHd3x939+qwT6hnYCH9vD+15JCIiYkfsZqbJzc2N0NBQYmNji43HxsbSu3fvUh8THh5eon7t2rWEhYXh6upaZs2VY1bkea+3K3sewf/2OLpCex6JiIjYht2EJoCYmBj+85//8P7775OUlMSkSZNISUkp2nfp6aefZsyYMUX1UVFRHD9+nJiYGJKSknj//feZP38+Tz75ZFHNxIkTWbt2LTNmzGD//v3MmDGDb7/9lujoaKuf1xau7Hnk5138VJ2ftwdzRoVozyMREZHqZtiZWbNmGa1atTLc3NyMkJAQY/369UX3jR071ujfv3+x+nXr1hk9evQw3NzcjNatWxtz5swpcczPPvvM6NChg+Hq6mp07NjRWLZsWbme1xpZWVkGYGRlZZXrcddSaLYYmw9nGl8mnDA2H840Cs2WKj2+iIhIbVaez2+72qepJqvKfZpERESkepTn89uuTs+JiIiI2CuFJhERERErKDSJiIiIWEGhSURERMQKCk0iIiIiVlBoEhEREbGCQpOIiIiIFRSaRERERKyg0CQiIiJiBRdbN+Aormysnp2dbeNORERExFpXPretuUCKQlMVycnJAaBFixY27kRERETKKycnB29v7zJrdO25KmKxWDh16hSenp6YTKYqPXZ2djYtWrQgNTVV17W7jvQ+Vw+9z9VD73P10Ptcfa7Xe20YBjk5OQQEBODkVPaqJc00VREnJyeaN29+XZ/Dy8tLfyirgd7n6qH3uXrofa4eep+rz/V4r681w3SFFoKLiIiIWEGhSURERMQKCk01gLu7O1OnTsXd3d3WrTg0vc/VQ+9z9dD7XD30Plcfe3ivtRBcRERExAqaaRIRERGxgkKTiIiIiBUUmkRERESsoNAkIiIiYgWFJjs3e/ZsAgMD8fDwIDQ0lI0bN9q6JYcyffp0brzxRjw9PWnatCnDhw/nwIEDtm7L4U2fPh2TyUR0dLStW3FIJ0+eZNSoUfj4+FC3bl26d+9OfHy8rdtyKIWFhTz33HMEBgZSp04d2rRpw0svvYTFYrF1azXahg0bGDp0KAEBAZhMJr788sti9xuGwYsvvkhAQAB16tThlltuYe/evdXWn0KTHVu6dCnR0dE8++yzJCQk0LdvXwYPHkxKSoqtW3MY69evZ/z48WzZsoXY2FgKCwuJiIjgwoULtm7NYf3000/MmzePrl272roVh3Tu3Dn69OmDq6sr33zzDfv27eONN96gQYMGtm7NocyYMYO5c+cyc+ZMkpKS+Mc//sE///lP3nnnHVu3VqNduHCBbt26MXPmzFLv/8c//sG///1vZs6cyU8//YSfnx8DBgwouv7rdWeI3erZs6cRFRVVbKxjx47GlClTbNSR48vIyDAAY/369bZuxSHl5OQY7dq1M2JjY43+/fsbEydOtHVLDmfy5MnGzTffbOs2HN4f/vAH4+GHHy42dvfddxujRo2yUUeOBzCWL19e9LPFYjH8/PyM119/vWgsNzfX8Pb2NubOnVstPWmmyU7l5+cTHx9PREREsfGIiAg2b95so64cX1ZWFgCNGjWycSeOafz48fzhD3/gjjvusHUrDmvlypWEhYVxzz330LRpU3r06MF7771n67Yczs0338x3333HwYMHAUhMTGTTpk0MGTLExp05ruTkZNLT04t9Lrq7u9O/f/9q+1zUBXvtVGZmJmazGV9f32Ljvr6+pKen26grx2YYBjExMdx8880EBwfbuh2Hs2TJEnbs2MFPP/1k61Yc2tGjR5kzZw4xMTE888wzbNu2jQkTJuDu7s6YMWNs3Z7DmDx5MllZWXTs2BFnZ2fMZjOvvvoq999/v61bc1hXPvtK+1w8fvx4tfSg0GTnTCZTsZ8NwygxJlXj8ccfZ9euXWzatMnWrTic1NRUJk6cyNq1a/Hw8LB1Ow7NYrEQFhbGa6+9BkCPHj3Yu3cvc+bMUWiqQkuXLmXRokV88skndO7cmZ07dxIdHU1AQABjx461dXsOzZafiwpNdqpx48Y4OzuXmFXKyMgokbKl8p544glWrlzJhg0baN68ua3bcTjx8fFkZGQQGhpaNGY2m9mwYQMzZ84kLy8PZ2dnG3boOPz9/QkKCio21qlTJ5YtW2ajjhzT3//+d6ZMmcJ9990HQJcuXTh+/DjTp09XaLpO/Pz8gMszTv7+/kXj1fm5qDVNdsrNzY3Q0FBiY2OLjcfGxtK7d28bdeV4DMPg8ccf54svvuD7778nMDDQ1i05pNtvv53du3ezc+fOoltYWBgjR45k586dCkxVqE+fPiW2zTh48CCtWrWyUUeO6eLFizg5Ff8IdXZ21pYD11FgYCB+fn7FPhfz8/NZv359tX0uaqbJjsXExDB69GjCwsIIDw9n3rx5pKSkEBUVZevWHMb48eP55JNPWLFiBZ6enkUze97e3tSpU8fG3TkOT0/PEuvE6tWrh4+Pj9aPVbFJkybRu3dvXnvtNe699162bdvGvHnzmDdvnq1bcyhDhw7l1VdfpWXLlnTu3JmEhAT+/e9/8/DDD9u6tRrt/PnzHD58uOjn5ORkdu7cSaNGjWjZsiXR0dG89tprtGvXjnbt2vHaa69Rt25dHnjggeppsFq+oycVNmvWLKNVq1aGm5ubERISoq/CVzGg1NuCBQts3ZrD05YD18+qVauM4OBgw93d3ejYsaMxb948W7fkcLKzs42JEycaLVu2NDw8PIw2bdoYzz77rJGXl2fr1mq0H374odTfyWPHjjUM4/K2A1OnTjX8/PwMd3d3o1+/fsbu3burrT+TYRhG9cQzERERkZpLa5pERERErKDQJCIiImIFhSYRERERKyg0iYiIiFhBoUlERETECgpNIiIiIlZQaBIRERGxgkKTiNR469atw2Qy8csvv9i6FRFxYApNIlLj3HLLLURHRxf93Lt3b9LS0vD29rZZTwpuIo5P154TkRrPzc2t6AroIiLXi2aaRKRGefDBB1m/fj1vvfUWJpMJk8nEBx98UGyW54MPPqBBgwZ89dVXdOjQgbp16/KnP/2JCxcu8OGHH9K6dWsaNmzIE088gdlsLjp2fn4+Tz31FM2aNaNevXr06tWLdevWFd1//Phxhg4dSsOGDalXrx6dO3dm9erVHDt2jFtvvRWAhg0bYjKZePDBBwEwDIN//OMftGnThjp16tCtWzc+//zzomNemaH6+uuv6datGx4eHvTq1Yvdu3df9/dSRMpHM00iUqO89dZbHDx4kODgYF566SUA9u7dW6Lu4sWLvP322yxZsoScnBzuvvtu7r77bho0aMDq1as5evQoI0aM4OabbyYyMhKAhx56iGPHjrFkyRICAgJYvnw5gwYNYvfu3bRr147x48eTn5/Phg0bqFevHvv27aN+/fq0aNGCZcuWMWLECA4cOICXlxd16tQB4LnnnuOLL75gzpw5tGvXjg0bNjBq1CiaNGlC//79i/r9+9//zltvvYWfnx/PPPMMd955JwcPHsTV1bUa3lURsUq1XRpYRKSK9O/f35g4cWLRz1eujH7u3DnDMAxjwYIFBmAcPny4qObRRx816tata+Tk5BSNDRw40Hj00UcNwzCMw4cPGyaTyTh58mSx57r99tuNp59+2jAMw+jSpYvx4osvltrT73swDMM4f/684eHhYWzevLlY7SOPPGLcf//9xR63ZMmSovvPnDlj1KlTx1i6dKmV74iIVAfNNImIQ6pbty5t27Yt+tnX15fWrVtTv379YmMZGRkA7NixA8MwaN++fbHj5OXl4ePjA8CECRN47LHHWLt2LXfccQcjRoyga9euV+1h37595ObmMmDAgGLj+fn59OjRo9hYeHh40b83atSIDh06kJSUVM5XLSLXk0KTiDik35/WMplMpY5ZLBYALBYLzs7OxMfH4+zsXKzuStD685//zMCBA/n6669Zu3Yt06dP54033uCJJ54otYcrx/76669p1qxZsfvc3d2v+RpMJtM1a0Sk+ig0iUiN4+bmVmwBd1Xo0aMHZrOZjIwM+vbte9W6Fi1aEBUVRVRUFE8//TTvvfceTzzxBG5ubgDF+goKCsLd3Z2UlJRi65dKs2XLFlq2bAnAuXPnOHjwIB07dqyCVyYiVUWhSURqnNatW7N161aOHTtG/fr1i2Z0KqN9+/aMHDmSMWPG8MYbb9CjRw8yMzP5/vvv6dKlC0OGDCE6OprBgwfTvn17zp07x/fff0+nTp0AaNWqFSaTia+++oohQ4ZQp04dPD09efLJJ5k0aRIWi4Wbb76Z7OxsNm/eTP369Rk7dmzR87/00kv4+Pjg6+vLs88+S+PGjRk+fHilX5eIVB1tOSAiNc6TTz6Js7MzQUFBNGnShJSUlCo57oIFCxgzZgx/+9vf6NChA3feeSdbt26lRYsWwOVZpPHjx9OpUycGDRpEhw4dmD17NgDNmjVj2rRpTJkyBV9fXx5//HEAXn75ZV544QWmT59Op06dGDhwIKtWrSIwMLDYc7/++utMnDiR0NBQ0tLSWLlyZdHslYjYB5NhGIatmxARqa3WrVvHrbfeyrlz52jQoIGt2xGRMmimSURERMQKCk0iIiIiVtDpOREREREraKZJRERExAoKTSIiIiJWUGgSERERsYJCk4iIiIgVFJpERERErKDQJCIiImIFhSYRERERKyg0iYiIiFhBoUlERETECv8Pg+hyw5wQ/pkAAAAASUVORK5CYII=",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -900,7 +1088,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -910,7 +1098,6 @@
     }
    ],
    "source": [
-    "\n",
     "for traj_id in traj_ids: \n",
     "    traj = db.get_trajectory_atoms(last_run_id, traj_id)\n",
     "    r0 = traj[0].get_positions()\n",
@@ -929,2713 +1116,37 @@
   {
    "cell_type": "code",
    "execution_count": 17,
-   "id": "b2e6f48d-3773-4b82-9801-c6d719e56dc8",
+   "id": "e64ad4c8-5e4c-4315-ad85-46963ac5a2fd",
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "text/plain": [
-       "[Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...)),\n",
-       " Atoms(symbols='CH68O32', pbc=True, cell=[10.499662478717669, 10.499662478717669, 10.499662478717669], momenta=..., calculator=SinglePointCalculator(...))]"
-      ]
-     },
-     "execution_count": 17,
-     "metadata": {},
-     "output_type": "execute_result"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "traj: 0, n_frames: 41\n",
+      "traj: 1, n_frames: 41\n",
+      "traj: 2, n_frames: 41\n",
+      "traj: 3, n_frames: 41\n",
+      "traj: 4, n_frames: 41\n",
+      "traj: 5, n_frames: 41\n",
+      "traj: 6, n_frames: 41\n",
+      "traj: 7, n_frames: 41\n",
+      "traj: 8, n_frames: 41\n",
+      "traj: 9, n_frames: 41\n",
+      "traj: 10, n_frames: 41\n",
+      "traj: 11, n_frames: 41\n",
+      "traj: 12, n_frames: 41\n",
+      "traj: 13, n_frames: 41\n",
+      "traj: 14, n_frames: 41\n",
+      "traj: 15, n_frames: 41\n"
+     ]
     }
    ],
    "source": [
-    "traj"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "id": "2718e392-a1c4-4032-a096-ed94bdd83ba6",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "events = db.list_chunk_events(last_run_id)\n",
-    "events = pd.DataFrame.from_records(events)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "id": "90b271f1-ed41-4cd3-95fa-65e245fcc8af",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "pd.set_option('display.max_rows', 500)"
+    "for traj_id in traj_ids: \n",
+    "    traj = db.get_trajectory_atoms(last_run_id, traj_id)\n",
+    "    print(f'traj: {traj_id}, n_frames: {len(traj)}')"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "id": "ba7e7a2a-6103-495b-adb0-0699e27dca8f",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>run_id</th>\n",
-       "      <th>traj_id</th>\n",
-       "      <th>chunk_id</th>\n",
-       "      <th>attempt_index</th>\n",
-       "      <th>event_type</th>\n",
-       "      <th>frame_id</th>\n",
-       "      <th>created_at</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:21:25.764049-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:11.366672-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:11.748947-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:13.074010-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:14.372854-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:15.596272-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>6</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11159.0</td>\n",
-       "      <td>2026-02-05 14:22:15.783566-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>7</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11159.0</td>\n",
-       "      <td>2026-02-05 14:22:16.501098-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>8</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11163.0</td>\n",
-       "      <td>2026-02-05 14:22:17.117685-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>9</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11163.0</td>\n",
-       "      <td>2026-02-05 14:22:17.814058-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>10</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11155.0</td>\n",
-       "      <td>2026-02-05 14:22:18.100677-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11155.0</td>\n",
-       "      <td>2026-02-05 14:22:18.393766-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>12</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11149.0</td>\n",
-       "      <td>2026-02-05 14:22:19.050518-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>13</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11149.0</td>\n",
-       "      <td>2026-02-05 14:22:20.049119-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>14</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11167.0</td>\n",
-       "      <td>2026-02-05 14:22:20.745443-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>15</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11167.0</td>\n",
-       "      <td>2026-02-05 14:22:21.378764-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>16</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:22.003596-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>17</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:22.097664-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>18</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:38.791024-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>19</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:08.253054-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>20</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:08.488313-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>21</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:09.291283-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>22</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:10.156855-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>23</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:20.763394-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>24</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:21.190235-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:21.879575-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>26</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:22.479311-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>27</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:22.659019-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>28</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11205.0</td>\n",
-       "      <td>2026-02-05 14:23:22.748933-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>29</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11205.0</td>\n",
-       "      <td>2026-02-05 14:23:23.015632-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>30</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11204.0</td>\n",
-       "      <td>2026-02-05 14:23:23.354079-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>31</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11204.0</td>\n",
-       "      <td>2026-02-05 14:23:23.686533-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>32</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11206.0</td>\n",
-       "      <td>2026-02-05 14:23:24.027205-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>33</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11206.0</td>\n",
-       "      <td>2026-02-05 14:23:24.291920-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>34</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11207.0</td>\n",
-       "      <td>2026-02-05 14:23:24.639631-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11207.0</td>\n",
-       "      <td>2026-02-05 14:23:24.913431-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>36</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11203.0</td>\n",
-       "      <td>2026-02-05 14:23:25.166486-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>37</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11203.0</td>\n",
-       "      <td>2026-02-05 14:23:25.429863-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>38</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:25.604115-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>39</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:25.774788-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>40</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:30.011577-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>41</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:40.593802-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>42</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:40.839608-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:41.405790-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>44</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>TRAJECTORY_COMPLETED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:41.501280-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>45</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:21:26.089299-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:15.073812-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>47</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:15.386242-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>48</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:16.587028-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>49</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:17.463712-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:26.992582-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>51</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:27.407418-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>52</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:28.093795-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>53</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>TRAJECTORY_COMPLETED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:28.338898-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>54</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:21:26.347353-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>55</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:18.490966-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>56</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:19.150310-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>57</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:20.508295-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>58</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:21.654949-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>59</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:22.271026-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>60</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11164.0</td>\n",
-       "      <td>2026-02-05 14:22:22.363871-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>61</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11164.0</td>\n",
-       "      <td>2026-02-05 14:22:22.631621-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>62</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11160.0</td>\n",
-       "      <td>2026-02-05 14:22:22.986289-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>63</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11160.0</td>\n",
-       "      <td>2026-02-05 14:22:23.351746-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>64</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11156.0</td>\n",
-       "      <td>2026-02-05 14:22:23.965671-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>65</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11156.0</td>\n",
-       "      <td>2026-02-05 14:22:24.510416-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>66</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11152.0</td>\n",
-       "      <td>2026-02-05 14:22:25.385093-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>67</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11152.0</td>\n",
-       "      <td>2026-02-05 14:22:25.881013-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>68</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11171.0</td>\n",
-       "      <td>2026-02-05 14:22:26.164285-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>69</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11171.0</td>\n",
-       "      <td>2026-02-05 14:22:26.436053-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>70</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:26.647289-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>71</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:26.993628-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>72</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:39.465910-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>73</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:14.710153-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>74</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:15.092748-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>75</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:15.670197-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>76</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:16.530196-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>77</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:16.789433-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>78</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11199.0</td>\n",
-       "      <td>2026-02-05 14:23:16.936698-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>79</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11199.0</td>\n",
-       "      <td>2026-02-05 14:23:17.210930-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>80</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11190.0</td>\n",
-       "      <td>2026-02-05 14:23:17.558167-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>81</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11190.0</td>\n",
-       "      <td>2026-02-05 14:23:17.917064-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>82</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11201.0</td>\n",
-       "      <td>2026-02-05 14:23:18.623927-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>83</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11201.0</td>\n",
-       "      <td>2026-02-05 14:23:19.257141-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>84</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11187.0</td>\n",
-       "      <td>2026-02-05 14:23:19.614476-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>85</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11187.0</td>\n",
-       "      <td>2026-02-05 14:23:20.231787-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>86</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11196.0</td>\n",
-       "      <td>2026-02-05 14:23:20.494303-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>87</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11196.0</td>\n",
-       "      <td>2026-02-05 14:23:21.021634-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>88</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:21.358106-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>89</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:21.530897-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>90</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:30.693905-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>91</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:41.806367-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>92</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:41.985480-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>93</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:42.536920-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>94</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:43.073850-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>95</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:43.262224-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>96</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11214.0</td>\n",
-       "      <td>2026-02-05 14:23:43.357090-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>97</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11214.0</td>\n",
-       "      <td>2026-02-05 14:23:43.706812-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>98</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11220.0</td>\n",
-       "      <td>2026-02-05 14:23:43.982586-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>99</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11220.0</td>\n",
-       "      <td>2026-02-05 14:23:44.248500-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>100</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11219.0</td>\n",
-       "      <td>2026-02-05 14:23:44.517229-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>101</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11219.0</td>\n",
-       "      <td>2026-02-05 14:23:44.884752-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>102</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11216.0</td>\n",
-       "      <td>2026-02-05 14:23:45.147274-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>103</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11216.0</td>\n",
-       "      <td>2026-02-05 14:23:45.411827-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>104</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11218.0</td>\n",
-       "      <td>2026-02-05 14:23:45.667503-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>105</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11218.0</td>\n",
-       "      <td>2026-02-05 14:23:46.023013-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>106</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:46.197611-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>107</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:46.291362-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>108</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:47.625240-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>109</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:49.666866-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>110</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:49.845306-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>111</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:50.297589-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>112</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:50.908108-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>113</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:51.089732-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>114</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11223.0</td>\n",
-       "      <td>2026-02-05 14:23:51.186062-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>115</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11223.0</td>\n",
-       "      <td>2026-02-05 14:23:51.443592-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>116</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11222.0</td>\n",
-       "      <td>2026-02-05 14:23:51.786857-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>117</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11222.0</td>\n",
-       "      <td>2026-02-05 14:23:52.050735-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>118</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11221.0</td>\n",
-       "      <td>2026-02-05 14:23:52.312944-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>119</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11221.0</td>\n",
-       "      <td>2026-02-05 14:23:52.582553-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>120</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11224.0</td>\n",
-       "      <td>2026-02-05 14:23:52.928478-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>121</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11224.0</td>\n",
-       "      <td>2026-02-05 14:23:53.360521-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>122</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11226.0</td>\n",
-       "      <td>2026-02-05 14:23:53.614335-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>123</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11226.0</td>\n",
-       "      <td>2026-02-05 14:23:53.966505-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>124</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:54.148304-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>125</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:54.242280-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>126</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:59.334506-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>127</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:01.346804-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>128</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:01.524135-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>129</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:02.068011-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>130</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:02.576737-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>131</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:04.635262-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>132</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:04.810762-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>133</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:05.522315-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>134</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:06.044906-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>135</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:06.238193-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>136</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11235.0</td>\n",
-       "      <td>2026-02-05 14:24:06.322763-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>137</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11235.0</td>\n",
-       "      <td>2026-02-05 14:24:06.681357-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>138</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11234.0</td>\n",
-       "      <td>2026-02-05 14:24:06.947617-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>139</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11234.0</td>\n",
-       "      <td>2026-02-05 14:24:07.224124-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>140</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11236.0</td>\n",
-       "      <td>2026-02-05 14:24:07.502809-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>141</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11236.0</td>\n",
-       "      <td>2026-02-05 14:24:07.852713-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>142</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11233.0</td>\n",
-       "      <td>2026-02-05 14:24:08.113719-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>143</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11233.0</td>\n",
-       "      <td>2026-02-05 14:24:08.390742-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>144</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11238.0</td>\n",
-       "      <td>2026-02-05 14:24:08.654090-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>145</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11238.0</td>\n",
-       "      <td>2026-02-05 14:24:09.014205-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>146</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:09.194523-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>147</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:09.288163-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>148</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:11.262536-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>149</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:13.338976-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>150</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:13.607045-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>151</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:14.069296-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>152</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:14.581237-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>153</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:14.852794-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>154</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11240.0</td>\n",
-       "      <td>2026-02-05 14:24:14.948347-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>155</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11240.0</td>\n",
-       "      <td>2026-02-05 14:24:15.217007-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>156</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11239.0</td>\n",
-       "      <td>2026-02-05 14:24:15.479073-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>157</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11239.0</td>\n",
-       "      <td>2026-02-05 14:24:15.746866-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>158</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11243.0</td>\n",
-       "      <td>2026-02-05 14:24:16.097318-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>159</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11243.0</td>\n",
-       "      <td>2026-02-05 14:24:16.374194-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>160</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11244.0</td>\n",
-       "      <td>2026-02-05 14:24:16.638587-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>161</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11244.0</td>\n",
-       "      <td>2026-02-05 14:24:17.076846-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>162</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11241.0</td>\n",
-       "      <td>2026-02-05 14:24:17.426998-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>163</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11241.0</td>\n",
-       "      <td>2026-02-05 14:24:17.699085-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>164</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:17.876246-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>165</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:17.973224-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>166</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:23.144571-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>167</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:25.214059-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>168</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:25.477769-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>169</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:25.917165-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>170</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:26.430163-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>171</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:26.696626-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>172</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11249.0</td>\n",
-       "      <td>2026-02-05 14:24:26.791634-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>173</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11249.0</td>\n",
-       "      <td>2026-02-05 14:24:27.063539-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>174</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11247.0</td>\n",
-       "      <td>2026-02-05 14:24:27.329985-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>175</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11247.0</td>\n",
-       "      <td>2026-02-05 14:24:27.600766-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>176</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11246.0</td>\n",
-       "      <td>2026-02-05 14:24:27.949107-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>177</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11246.0</td>\n",
-       "      <td>2026-02-05 14:24:28.231915-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>178</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11248.0</td>\n",
-       "      <td>2026-02-05 14:24:28.496793-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>179</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11248.0</td>\n",
-       "      <td>2026-02-05 14:24:28.930931-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>180</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11245.0</td>\n",
-       "      <td>2026-02-05 14:24:29.285036-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>181</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11245.0</td>\n",
-       "      <td>2026-02-05 14:24:29.565742-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>182</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:29.741568-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>183</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:29.835286-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>184</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:34.970079-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>185</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:36.911578-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>186</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:37.093841-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>187</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:37.622168-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>188</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>TRAJECTORY_COMPLETED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:24:37.798723-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>189</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:21:26.786171-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>190</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:23.252417-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>191</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:23.606524-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>192</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:24.593704-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>193</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:25.385513-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>194</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:28.849906-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>195</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:29.029835-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>196</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>AUDIT_FAILED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:29.508604-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>197</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:30.116614-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>198</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:30.304135-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>199</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11179.0</td>\n",
-       "      <td>2026-02-05 14:22:30.398043-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>200</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11179.0</td>\n",
-       "      <td>2026-02-05 14:22:30.683030-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>201</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11180.0</td>\n",
-       "      <td>2026-02-05 14:22:31.059885-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>202</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11180.0</td>\n",
-       "      <td>2026-02-05 14:22:31.330180-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>203</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11184.0</td>\n",
-       "      <td>2026-02-05 14:22:31.593516-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>204</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11184.0</td>\n",
-       "      <td>2026-02-05 14:22:31.847269-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>205</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11181.0</td>\n",
-       "      <td>2026-02-05 14:22:32.232547-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>206</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11181.0</td>\n",
-       "      <td>2026-02-05 14:22:32.657863-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>207</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>STARTED_LABELING_FRAME</td>\n",
-       "      <td>11183.0</td>\n",
-       "      <td>2026-02-05 14:22:32.952476-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>208</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING_FRAME</td>\n",
-       "      <td>11183.0</td>\n",
-       "      <td>2026-02-05 14:22:33.227929-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>209</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_LABELING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:33.523944-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>210</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>FINISHED_SAMPLING</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:33.622367-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>211</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:22:40.222113-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>212</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>FINISHED_DYNAMICS</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:17.916817-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>213</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>STARTED_AUDIT</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:18.272984-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>214</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>AUDIT_PASSED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:19.613677-06:00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>215</th>\n",
-       "      <td>2026.02.05-20:21:24-c0e31c</td>\n",
-       "      <td>3</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>TRAJECTORY_COMPLETED</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>2026-02-05 14:23:19.959950-06:00</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                         run_id  traj_id  chunk_id  attempt_index  \\\n",
-       "0    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "1    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "2    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "3    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "4    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "5    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "6    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "7    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "8    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "9    2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "10   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "11   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "12   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "13   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "14   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "15   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "16   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "17   2026.02.05-20:21:24-c0e31c        0         0              0   \n",
-       "18   2026.02.05-20:21:24-c0e31c        0         0              1   \n",
-       "19   2026.02.05-20:21:24-c0e31c        0         0              1   \n",
-       "20   2026.02.05-20:21:24-c0e31c        0         0              1   \n",
-       "21   2026.02.05-20:21:24-c0e31c        0         0              1   \n",
-       "22   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "23   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "24   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "25   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "26   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "27   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "28   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "29   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "30   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "31   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "32   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "33   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "34   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "35   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "36   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "37   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "38   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "39   2026.02.05-20:21:24-c0e31c        0         1              0   \n",
-       "40   2026.02.05-20:21:24-c0e31c        0         1              1   \n",
-       "41   2026.02.05-20:21:24-c0e31c        0         1              1   \n",
-       "42   2026.02.05-20:21:24-c0e31c        0         1              1   \n",
-       "43   2026.02.05-20:21:24-c0e31c        0         1              1   \n",
-       "44   2026.02.05-20:21:24-c0e31c        0         1              1   \n",
-       "45   2026.02.05-20:21:24-c0e31c        1         0              0   \n",
-       "46   2026.02.05-20:21:24-c0e31c        1         0              0   \n",
-       "47   2026.02.05-20:21:24-c0e31c        1         0              0   \n",
-       "48   2026.02.05-20:21:24-c0e31c        1         0              0   \n",
-       "49   2026.02.05-20:21:24-c0e31c        1         1              0   \n",
-       "50   2026.02.05-20:21:24-c0e31c        1         1              0   \n",
-       "51   2026.02.05-20:21:24-c0e31c        1         1              0   \n",
-       "52   2026.02.05-20:21:24-c0e31c        1         1              0   \n",
-       "53   2026.02.05-20:21:24-c0e31c        1         1              0   \n",
-       "54   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "55   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "56   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "57   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "58   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "59   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "60   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "61   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "62   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "63   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "64   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "65   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "66   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "67   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "68   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "69   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "70   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "71   2026.02.05-20:21:24-c0e31c        2         0              0   \n",
-       "72   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "73   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "74   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "75   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "76   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "77   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "78   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "79   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "80   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "81   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "82   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "83   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "84   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "85   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "86   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "87   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "88   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "89   2026.02.05-20:21:24-c0e31c        2         0              1   \n",
-       "90   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "91   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "92   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "93   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "94   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "95   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "96   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "97   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "98   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "99   2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "100  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "101  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "102  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "103  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "104  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "105  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "106  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "107  2026.02.05-20:21:24-c0e31c        2         0              2   \n",
-       "108  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "109  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "110  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "111  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "112  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "113  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "114  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "115  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "116  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "117  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "118  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "119  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "120  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "121  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "122  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "123  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "124  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "125  2026.02.05-20:21:24-c0e31c        2         0              3   \n",
-       "126  2026.02.05-20:21:24-c0e31c        2         0              4   \n",
-       "127  2026.02.05-20:21:24-c0e31c        2         0              4   \n",
-       "128  2026.02.05-20:21:24-c0e31c        2         0              4   \n",
-       "129  2026.02.05-20:21:24-c0e31c        2         0              4   \n",
-       "130  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "131  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "132  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "133  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "134  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "135  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "136  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "137  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "138  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "139  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "140  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "141  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "142  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "143  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "144  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "145  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "146  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "147  2026.02.05-20:21:24-c0e31c        2         1              0   \n",
-       "148  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "149  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "150  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "151  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "152  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "153  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "154  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "155  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "156  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "157  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "158  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "159  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "160  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "161  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "162  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "163  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "164  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "165  2026.02.05-20:21:24-c0e31c        2         1              1   \n",
-       "166  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "167  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "168  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "169  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "170  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "171  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "172  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "173  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "174  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "175  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "176  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "177  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "178  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "179  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "180  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "181  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "182  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "183  2026.02.05-20:21:24-c0e31c        2         1              2   \n",
-       "184  2026.02.05-20:21:24-c0e31c        2         1              3   \n",
-       "185  2026.02.05-20:21:24-c0e31c        2         1              3   \n",
-       "186  2026.02.05-20:21:24-c0e31c        2         1              3   \n",
-       "187  2026.02.05-20:21:24-c0e31c        2         1              3   \n",
-       "188  2026.02.05-20:21:24-c0e31c        2         1              3   \n",
-       "189  2026.02.05-20:21:24-c0e31c        3         0              0   \n",
-       "190  2026.02.05-20:21:24-c0e31c        3         0              0   \n",
-       "191  2026.02.05-20:21:24-c0e31c        3         0              0   \n",
-       "192  2026.02.05-20:21:24-c0e31c        3         0              0   \n",
-       "193  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "194  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "195  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "196  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "197  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "198  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "199  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "200  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "201  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "202  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "203  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "204  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "205  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "206  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "207  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "208  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "209  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "210  2026.02.05-20:21:24-c0e31c        3         1              0   \n",
-       "211  2026.02.05-20:21:24-c0e31c        3         1              1   \n",
-       "212  2026.02.05-20:21:24-c0e31c        3         1              1   \n",
-       "213  2026.02.05-20:21:24-c0e31c        3         1              1   \n",
-       "214  2026.02.05-20:21:24-c0e31c        3         1              1   \n",
-       "215  2026.02.05-20:21:24-c0e31c        3         1              1   \n",
-       "\n",
-       "                  event_type  frame_id                       created_at  \n",
-       "0           STARTED_DYNAMICS       NaN 2026-02-05 14:21:25.764049-06:00  \n",
-       "1          FINISHED_DYNAMICS       NaN 2026-02-05 14:22:11.366672-06:00  \n",
-       "2              STARTED_AUDIT       NaN 2026-02-05 14:22:11.748947-06:00  \n",
-       "3               AUDIT_FAILED       NaN 2026-02-05 14:22:13.074010-06:00  \n",
-       "4           STARTED_SAMPLING       NaN 2026-02-05 14:22:14.372854-06:00  \n",
-       "5           STARTED_LABELING       NaN 2026-02-05 14:22:15.596272-06:00  \n",
-       "6     STARTED_LABELING_FRAME   11159.0 2026-02-05 14:22:15.783566-06:00  \n",
-       "7    FINISHED_LABELING_FRAME   11159.0 2026-02-05 14:22:16.501098-06:00  \n",
-       "8     STARTED_LABELING_FRAME   11163.0 2026-02-05 14:22:17.117685-06:00  \n",
-       "9    FINISHED_LABELING_FRAME   11163.0 2026-02-05 14:22:17.814058-06:00  \n",
-       "10    STARTED_LABELING_FRAME   11155.0 2026-02-05 14:22:18.100677-06:00  \n",
-       "11   FINISHED_LABELING_FRAME   11155.0 2026-02-05 14:22:18.393766-06:00  \n",
-       "12    STARTED_LABELING_FRAME   11149.0 2026-02-05 14:22:19.050518-06:00  \n",
-       "13   FINISHED_LABELING_FRAME   11149.0 2026-02-05 14:22:20.049119-06:00  \n",
-       "14    STARTED_LABELING_FRAME   11167.0 2026-02-05 14:22:20.745443-06:00  \n",
-       "15   FINISHED_LABELING_FRAME   11167.0 2026-02-05 14:22:21.378764-06:00  \n",
-       "16         FINISHED_LABELING       NaN 2026-02-05 14:22:22.003596-06:00  \n",
-       "17         FINISHED_SAMPLING       NaN 2026-02-05 14:22:22.097664-06:00  \n",
-       "18          STARTED_DYNAMICS       NaN 2026-02-05 14:22:38.791024-06:00  \n",
-       "19         FINISHED_DYNAMICS       NaN 2026-02-05 14:23:08.253054-06:00  \n",
-       "20             STARTED_AUDIT       NaN 2026-02-05 14:23:08.488313-06:00  \n",
-       "21              AUDIT_PASSED       NaN 2026-02-05 14:23:09.291283-06:00  \n",
-       "22          STARTED_DYNAMICS       NaN 2026-02-05 14:23:10.156855-06:00  \n",
-       "23         FINISHED_DYNAMICS       NaN 2026-02-05 14:23:20.763394-06:00  \n",
-       "24             STARTED_AUDIT       NaN 2026-02-05 14:23:21.190235-06:00  \n",
-       "25              AUDIT_FAILED       NaN 2026-02-05 14:23:21.879575-06:00  \n",
-       "26          STARTED_SAMPLING       NaN 2026-02-05 14:23:22.479311-06:00  \n",
-       "27          STARTED_LABELING       NaN 2026-02-05 14:23:22.659019-06:00  \n",
-       "28    STARTED_LABELING_FRAME   11205.0 2026-02-05 14:23:22.748933-06:00  \n",
-       "29   FINISHED_LABELING_FRAME   11205.0 2026-02-05 14:23:23.015632-06:00  \n",
-       "30    STARTED_LABELING_FRAME   11204.0 2026-02-05 14:23:23.354079-06:00  \n",
-       "31   FINISHED_LABELING_FRAME   11204.0 2026-02-05 14:23:23.686533-06:00  \n",
-       "32    STARTED_LABELING_FRAME   11206.0 2026-02-05 14:23:24.027205-06:00  \n",
-       "33   FINISHED_LABELING_FRAME   11206.0 2026-02-05 14:23:24.291920-06:00  \n",
-       "34    STARTED_LABELING_FRAME   11207.0 2026-02-05 14:23:24.639631-06:00  \n",
-       "35   FINISHED_LABELING_FRAME   11207.0 2026-02-05 14:23:24.913431-06:00  \n",
-       "36    STARTED_LABELING_FRAME   11203.0 2026-02-05 14:23:25.166486-06:00  \n",
-       "37   FINISHED_LABELING_FRAME   11203.0 2026-02-05 14:23:25.429863-06:00  \n",
-       "38         FINISHED_LABELING       NaN 2026-02-05 14:23:25.604115-06:00  \n",
-       "39         FINISHED_SAMPLING       NaN 2026-02-05 14:23:25.774788-06:00  \n",
-       "40          STARTED_DYNAMICS       NaN 2026-02-05 14:23:30.011577-06:00  \n",
-       "41         FINISHED_DYNAMICS       NaN 2026-02-05 14:23:40.593802-06:00  \n",
-       "42             STARTED_AUDIT       NaN 2026-02-05 14:23:40.839608-06:00  \n",
-       "43              AUDIT_PASSED       NaN 2026-02-05 14:23:41.405790-06:00  \n",
-       "44      TRAJECTORY_COMPLETED       NaN 2026-02-05 14:23:41.501280-06:00  \n",
-       "45          STARTED_DYNAMICS       NaN 2026-02-05 14:21:26.089299-06:00  \n",
-       "46         FINISHED_DYNAMICS       NaN 2026-02-05 14:22:15.073812-06:00  \n",
-       "47             STARTED_AUDIT       NaN 2026-02-05 14:22:15.386242-06:00  \n",
-       "48              AUDIT_PASSED       NaN 2026-02-05 14:22:16.587028-06:00  \n",
-       "49          STARTED_DYNAMICS       NaN 2026-02-05 14:22:17.463712-06:00  \n",
-       "50         FINISHED_DYNAMICS       NaN 2026-02-05 14:22:26.992582-06:00  \n",
-       "51             STARTED_AUDIT       NaN 2026-02-05 14:22:27.407418-06:00  \n",
-       "52              AUDIT_PASSED       NaN 2026-02-05 14:22:28.093795-06:00  \n",
-       "53      TRAJECTORY_COMPLETED       NaN 2026-02-05 14:22:28.338898-06:00  \n",
-       "54          STARTED_DYNAMICS       NaN 2026-02-05 14:21:26.347353-06:00  \n",
-       "55         FINISHED_DYNAMICS       NaN 2026-02-05 14:22:18.490966-06:00  \n",
-       "56             STARTED_AUDIT       NaN 2026-02-05 14:22:19.150310-06:00  \n",
-       "57              AUDIT_FAILED       NaN 2026-02-05 14:22:20.508295-06:00  \n",
-       "58          STARTED_SAMPLING       NaN 2026-02-05 14:22:21.654949-06:00  \n",
-       "59          STARTED_LABELING       NaN 2026-02-05 14:22:22.271026-06:00  \n",
-       "60    STARTED_LABELING_FRAME   11164.0 2026-02-05 14:22:22.363871-06:00  \n",
-       "61   FINISHED_LABELING_FRAME   11164.0 2026-02-05 14:22:22.631621-06:00  \n",
-       "62    STARTED_LABELING_FRAME   11160.0 2026-02-05 14:22:22.986289-06:00  \n",
-       "63   FINISHED_LABELING_FRAME   11160.0 2026-02-05 14:22:23.351746-06:00  \n",
-       "64    STARTED_LABELING_FRAME   11156.0 2026-02-05 14:22:23.965671-06:00  \n",
-       "65   FINISHED_LABELING_FRAME   11156.0 2026-02-05 14:22:24.510416-06:00  \n",
-       "66    STARTED_LABELING_FRAME   11152.0 2026-02-05 14:22:25.385093-06:00  \n",
-       "67   FINISHED_LABELING_FRAME   11152.0 2026-02-05 14:22:25.881013-06:00  \n",
-       "68    STARTED_LABELING_FRAME   11171.0 2026-02-05 14:22:26.164285-06:00  \n",
-       "69   FINISHED_LABELING_FRAME   11171.0 2026-02-05 14:22:26.436053-06:00  \n",
-       "70         FINISHED_LABELING       NaN 2026-02-05 14:22:26.647289-06:00  \n",
-       "71         FINISHED_SAMPLING       NaN 2026-02-05 14:22:26.993628-06:00  \n",
-       "72          STARTED_DYNAMICS       NaN 2026-02-05 14:22:39.465910-06:00  \n",
-       "73         FINISHED_DYNAMICS       NaN 2026-02-05 14:23:14.710153-06:00  \n",
-       "74             STARTED_AUDIT       NaN 2026-02-05 14:23:15.092748-06:00  \n",
-       "75              AUDIT_FAILED       NaN 2026-02-05 14:23:15.670197-06:00  \n",
-       "76          STARTED_SAMPLING       NaN 2026-02-05 14:23:16.530196-06:00  \n",
-       "77          STARTED_LABELING       NaN 2026-02-05 14:23:16.789433-06:00  \n",
-       "78    STARTED_LABELING_FRAME   11199.0 2026-02-05 14:23:16.936698-06:00  \n",
-       "79   FINISHED_LABELING_FRAME   11199.0 2026-02-05 14:23:17.210930-06:00  \n",
-       "80    STARTED_LABELING_FRAME   11190.0 2026-02-05 14:23:17.558167-06:00  \n",
-       "81   FINISHED_LABELING_FRAME   11190.0 2026-02-05 14:23:17.917064-06:00  \n",
-       "82    STARTED_LABELING_FRAME   11201.0 2026-02-05 14:23:18.623927-06:00  \n",
-       "83   FINISHED_LABELING_FRAME   11201.0 2026-02-05 14:23:19.257141-06:00  \n",
-       "84    STARTED_LABELING_FRAME   11187.0 2026-02-05 14:23:19.614476-06:00  \n",
-       "85   FINISHED_LABELING_FRAME   11187.0 2026-02-05 14:23:20.231787-06:00  \n",
-       "86    STARTED_LABELING_FRAME   11196.0 2026-02-05 14:23:20.494303-06:00  \n",
-       "87   FINISHED_LABELING_FRAME   11196.0 2026-02-05 14:23:21.021634-06:00  \n",
-       "88         FINISHED_LABELING       NaN 2026-02-05 14:23:21.358106-06:00  \n",
-       "89         FINISHED_SAMPLING       NaN 2026-02-05 14:23:21.530897-06:00  \n",
-       "90          STARTED_DYNAMICS       NaN 2026-02-05 14:23:30.693905-06:00  \n",
-       "91         FINISHED_DYNAMICS       NaN 2026-02-05 14:23:41.806367-06:00  \n",
-       "92             STARTED_AUDIT       NaN 2026-02-05 14:23:41.985480-06:00  \n",
-       "93              AUDIT_FAILED       NaN 2026-02-05 14:23:42.536920-06:00  \n",
-       "94          STARTED_SAMPLING       NaN 2026-02-05 14:23:43.073850-06:00  \n",
-       "95          STARTED_LABELING       NaN 2026-02-05 14:23:43.262224-06:00  \n",
-       "96    STARTED_LABELING_FRAME   11214.0 2026-02-05 14:23:43.357090-06:00  \n",
-       "97   FINISHED_LABELING_FRAME   11214.0 2026-02-05 14:23:43.706812-06:00  \n",
-       "98    STARTED_LABELING_FRAME   11220.0 2026-02-05 14:23:43.982586-06:00  \n",
-       "99   FINISHED_LABELING_FRAME   11220.0 2026-02-05 14:23:44.248500-06:00  \n",
-       "100   STARTED_LABELING_FRAME   11219.0 2026-02-05 14:23:44.517229-06:00  \n",
-       "101  FINISHED_LABELING_FRAME   11219.0 2026-02-05 14:23:44.884752-06:00  \n",
-       "102   STARTED_LABELING_FRAME   11216.0 2026-02-05 14:23:45.147274-06:00  \n",
-       "103  FINISHED_LABELING_FRAME   11216.0 2026-02-05 14:23:45.411827-06:00  \n",
-       "104   STARTED_LABELING_FRAME   11218.0 2026-02-05 14:23:45.667503-06:00  \n",
-       "105  FINISHED_LABELING_FRAME   11218.0 2026-02-05 14:23:46.023013-06:00  \n",
-       "106        FINISHED_LABELING       NaN 2026-02-05 14:23:46.197611-06:00  \n",
-       "107        FINISHED_SAMPLING       NaN 2026-02-05 14:23:46.291362-06:00  \n",
-       "108         STARTED_DYNAMICS       NaN 2026-02-05 14:23:47.625240-06:00  \n",
-       "109        FINISHED_DYNAMICS       NaN 2026-02-05 14:23:49.666866-06:00  \n",
-       "110            STARTED_AUDIT       NaN 2026-02-05 14:23:49.845306-06:00  \n",
-       "111             AUDIT_FAILED       NaN 2026-02-05 14:23:50.297589-06:00  \n",
-       "112         STARTED_SAMPLING       NaN 2026-02-05 14:23:50.908108-06:00  \n",
-       "113         STARTED_LABELING       NaN 2026-02-05 14:23:51.089732-06:00  \n",
-       "114   STARTED_LABELING_FRAME   11223.0 2026-02-05 14:23:51.186062-06:00  \n",
-       "115  FINISHED_LABELING_FRAME   11223.0 2026-02-05 14:23:51.443592-06:00  \n",
-       "116   STARTED_LABELING_FRAME   11222.0 2026-02-05 14:23:51.786857-06:00  \n",
-       "117  FINISHED_LABELING_FRAME   11222.0 2026-02-05 14:23:52.050735-06:00  \n",
-       "118   STARTED_LABELING_FRAME   11221.0 2026-02-05 14:23:52.312944-06:00  \n",
-       "119  FINISHED_LABELING_FRAME   11221.0 2026-02-05 14:23:52.582553-06:00  \n",
-       "120   STARTED_LABELING_FRAME   11224.0 2026-02-05 14:23:52.928478-06:00  \n",
-       "121  FINISHED_LABELING_FRAME   11224.0 2026-02-05 14:23:53.360521-06:00  \n",
-       "122   STARTED_LABELING_FRAME   11226.0 2026-02-05 14:23:53.614335-06:00  \n",
-       "123  FINISHED_LABELING_FRAME   11226.0 2026-02-05 14:23:53.966505-06:00  \n",
-       "124        FINISHED_LABELING       NaN 2026-02-05 14:23:54.148304-06:00  \n",
-       "125        FINISHED_SAMPLING       NaN 2026-02-05 14:23:54.242280-06:00  \n",
-       "126         STARTED_DYNAMICS       NaN 2026-02-05 14:23:59.334506-06:00  \n",
-       "127        FINISHED_DYNAMICS       NaN 2026-02-05 14:24:01.346804-06:00  \n",
-       "128            STARTED_AUDIT       NaN 2026-02-05 14:24:01.524135-06:00  \n",
-       "129             AUDIT_PASSED       NaN 2026-02-05 14:24:02.068011-06:00  \n",
-       "130         STARTED_DYNAMICS       NaN 2026-02-05 14:24:02.576737-06:00  \n",
-       "131        FINISHED_DYNAMICS       NaN 2026-02-05 14:24:04.635262-06:00  \n",
-       "132            STARTED_AUDIT       NaN 2026-02-05 14:24:04.810762-06:00  \n",
-       "133             AUDIT_FAILED       NaN 2026-02-05 14:24:05.522315-06:00  \n",
-       "134         STARTED_SAMPLING       NaN 2026-02-05 14:24:06.044906-06:00  \n",
-       "135         STARTED_LABELING       NaN 2026-02-05 14:24:06.238193-06:00  \n",
-       "136   STARTED_LABELING_FRAME   11235.0 2026-02-05 14:24:06.322763-06:00  \n",
-       "137  FINISHED_LABELING_FRAME   11235.0 2026-02-05 14:24:06.681357-06:00  \n",
-       "138   STARTED_LABELING_FRAME   11234.0 2026-02-05 14:24:06.947617-06:00  \n",
-       "139  FINISHED_LABELING_FRAME   11234.0 2026-02-05 14:24:07.224124-06:00  \n",
-       "140   STARTED_LABELING_FRAME   11236.0 2026-02-05 14:24:07.502809-06:00  \n",
-       "141  FINISHED_LABELING_FRAME   11236.0 2026-02-05 14:24:07.852713-06:00  \n",
-       "142   STARTED_LABELING_FRAME   11233.0 2026-02-05 14:24:08.113719-06:00  \n",
-       "143  FINISHED_LABELING_FRAME   11233.0 2026-02-05 14:24:08.390742-06:00  \n",
-       "144   STARTED_LABELING_FRAME   11238.0 2026-02-05 14:24:08.654090-06:00  \n",
-       "145  FINISHED_LABELING_FRAME   11238.0 2026-02-05 14:24:09.014205-06:00  \n",
-       "146        FINISHED_LABELING       NaN 2026-02-05 14:24:09.194523-06:00  \n",
-       "147        FINISHED_SAMPLING       NaN 2026-02-05 14:24:09.288163-06:00  \n",
-       "148         STARTED_DYNAMICS       NaN 2026-02-05 14:24:11.262536-06:00  \n",
-       "149        FINISHED_DYNAMICS       NaN 2026-02-05 14:24:13.338976-06:00  \n",
-       "150            STARTED_AUDIT       NaN 2026-02-05 14:24:13.607045-06:00  \n",
-       "151             AUDIT_FAILED       NaN 2026-02-05 14:24:14.069296-06:00  \n",
-       "152         STARTED_SAMPLING       NaN 2026-02-05 14:24:14.581237-06:00  \n",
-       "153         STARTED_LABELING       NaN 2026-02-05 14:24:14.852794-06:00  \n",
-       "154   STARTED_LABELING_FRAME   11240.0 2026-02-05 14:24:14.948347-06:00  \n",
-       "155  FINISHED_LABELING_FRAME   11240.0 2026-02-05 14:24:15.217007-06:00  \n",
-       "156   STARTED_LABELING_FRAME   11239.0 2026-02-05 14:24:15.479073-06:00  \n",
-       "157  FINISHED_LABELING_FRAME   11239.0 2026-02-05 14:24:15.746866-06:00  \n",
-       "158   STARTED_LABELING_FRAME   11243.0 2026-02-05 14:24:16.097318-06:00  \n",
-       "159  FINISHED_LABELING_FRAME   11243.0 2026-02-05 14:24:16.374194-06:00  \n",
-       "160   STARTED_LABELING_FRAME   11244.0 2026-02-05 14:24:16.638587-06:00  \n",
-       "161  FINISHED_LABELING_FRAME   11244.0 2026-02-05 14:24:17.076846-06:00  \n",
-       "162   STARTED_LABELING_FRAME   11241.0 2026-02-05 14:24:17.426998-06:00  \n",
-       "163  FINISHED_LABELING_FRAME   11241.0 2026-02-05 14:24:17.699085-06:00  \n",
-       "164        FINISHED_LABELING       NaN 2026-02-05 14:24:17.876246-06:00  \n",
-       "165        FINISHED_SAMPLING       NaN 2026-02-05 14:24:17.973224-06:00  \n",
-       "166         STARTED_DYNAMICS       NaN 2026-02-05 14:24:23.144571-06:00  \n",
-       "167        FINISHED_DYNAMICS       NaN 2026-02-05 14:24:25.214059-06:00  \n",
-       "168            STARTED_AUDIT       NaN 2026-02-05 14:24:25.477769-06:00  \n",
-       "169             AUDIT_FAILED       NaN 2026-02-05 14:24:25.917165-06:00  \n",
-       "170         STARTED_SAMPLING       NaN 2026-02-05 14:24:26.430163-06:00  \n",
-       "171         STARTED_LABELING       NaN 2026-02-05 14:24:26.696626-06:00  \n",
-       "172   STARTED_LABELING_FRAME   11249.0 2026-02-05 14:24:26.791634-06:00  \n",
-       "173  FINISHED_LABELING_FRAME   11249.0 2026-02-05 14:24:27.063539-06:00  \n",
-       "174   STARTED_LABELING_FRAME   11247.0 2026-02-05 14:24:27.329985-06:00  \n",
-       "175  FINISHED_LABELING_FRAME   11247.0 2026-02-05 14:24:27.600766-06:00  \n",
-       "176   STARTED_LABELING_FRAME   11246.0 2026-02-05 14:24:27.949107-06:00  \n",
-       "177  FINISHED_LABELING_FRAME   11246.0 2026-02-05 14:24:28.231915-06:00  \n",
-       "178   STARTED_LABELING_FRAME   11248.0 2026-02-05 14:24:28.496793-06:00  \n",
-       "179  FINISHED_LABELING_FRAME   11248.0 2026-02-05 14:24:28.930931-06:00  \n",
-       "180   STARTED_LABELING_FRAME   11245.0 2026-02-05 14:24:29.285036-06:00  \n",
-       "181  FINISHED_LABELING_FRAME   11245.0 2026-02-05 14:24:29.565742-06:00  \n",
-       "182        FINISHED_LABELING       NaN 2026-02-05 14:24:29.741568-06:00  \n",
-       "183        FINISHED_SAMPLING       NaN 2026-02-05 14:24:29.835286-06:00  \n",
-       "184         STARTED_DYNAMICS       NaN 2026-02-05 14:24:34.970079-06:00  \n",
-       "185        FINISHED_DYNAMICS       NaN 2026-02-05 14:24:36.911578-06:00  \n",
-       "186            STARTED_AUDIT       NaN 2026-02-05 14:24:37.093841-06:00  \n",
-       "187             AUDIT_PASSED       NaN 2026-02-05 14:24:37.622168-06:00  \n",
-       "188     TRAJECTORY_COMPLETED       NaN 2026-02-05 14:24:37.798723-06:00  \n",
-       "189         STARTED_DYNAMICS       NaN 2026-02-05 14:21:26.786171-06:00  \n",
-       "190        FINISHED_DYNAMICS       NaN 2026-02-05 14:22:23.252417-06:00  \n",
-       "191            STARTED_AUDIT       NaN 2026-02-05 14:22:23.606524-06:00  \n",
-       "192             AUDIT_PASSED       NaN 2026-02-05 14:22:24.593704-06:00  \n",
-       "193         STARTED_DYNAMICS       NaN 2026-02-05 14:22:25.385513-06:00  \n",
-       "194        FINISHED_DYNAMICS       NaN 2026-02-05 14:22:28.849906-06:00  \n",
-       "195            STARTED_AUDIT       NaN 2026-02-05 14:22:29.029835-06:00  \n",
-       "196             AUDIT_FAILED       NaN 2026-02-05 14:22:29.508604-06:00  \n",
-       "197         STARTED_SAMPLING       NaN 2026-02-05 14:22:30.116614-06:00  \n",
-       "198         STARTED_LABELING       NaN 2026-02-05 14:22:30.304135-06:00  \n",
-       "199   STARTED_LABELING_FRAME   11179.0 2026-02-05 14:22:30.398043-06:00  \n",
-       "200  FINISHED_LABELING_FRAME   11179.0 2026-02-05 14:22:30.683030-06:00  \n",
-       "201   STARTED_LABELING_FRAME   11180.0 2026-02-05 14:22:31.059885-06:00  \n",
-       "202  FINISHED_LABELING_FRAME   11180.0 2026-02-05 14:22:31.330180-06:00  \n",
-       "203   STARTED_LABELING_FRAME   11184.0 2026-02-05 14:22:31.593516-06:00  \n",
-       "204  FINISHED_LABELING_FRAME   11184.0 2026-02-05 14:22:31.847269-06:00  \n",
-       "205   STARTED_LABELING_FRAME   11181.0 2026-02-05 14:22:32.232547-06:00  \n",
-       "206  FINISHED_LABELING_FRAME   11181.0 2026-02-05 14:22:32.657863-06:00  \n",
-       "207   STARTED_LABELING_FRAME   11183.0 2026-02-05 14:22:32.952476-06:00  \n",
-       "208  FINISHED_LABELING_FRAME   11183.0 2026-02-05 14:22:33.227929-06:00  \n",
-       "209        FINISHED_LABELING       NaN 2026-02-05 14:22:33.523944-06:00  \n",
-       "210        FINISHED_SAMPLING       NaN 2026-02-05 14:22:33.622367-06:00  \n",
-       "211         STARTED_DYNAMICS       NaN 2026-02-05 14:22:40.222113-06:00  \n",
-       "212        FINISHED_DYNAMICS       NaN 2026-02-05 14:23:17.916817-06:00  \n",
-       "213            STARTED_AUDIT       NaN 2026-02-05 14:23:18.272984-06:00  \n",
-       "214             AUDIT_PASSED       NaN 2026-02-05 14:23:19.613677-06:00  \n",
-       "215     TRAJECTORY_COMPLETED       NaN 2026-02-05 14:23:19.959950-06:00  "
-      ]
-     },
-     "execution_count": 21,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "events.sort_values([\n",
-    "    'traj_id', \n",
-    "    'chunk_id', \n",
-    "    'created_at'])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "d2be97e8-9f73-4c3b-a657-f57e4bdd9065",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {
diff --git a/3_cascade/run_cascade_academy.py b/3_cascade/run_cascade_academy.py
index b7c5547..37d21e1 100644
--- a/3_cascade/run_cascade_academy.py
+++ b/3_cascade/run_cascade_academy.py
@@ -7,32 +7,34 @@
 import hashlib
 import json
 import pathlib
+import sys
 
 import ase
 from ase.io import read
 from ase import units
 from ase.md.verlet import VelocityVerlet
-
-# Suppress FutureWarning about torch.load weights_only parameter from MACE
-warnings.filterwarnings("ignore", category=FutureWarning, module="mace.calculators")
-
 from mace.calculators import mace_mp
+from parsl.config import Config
+from parsl.executors import HighThroughputExecutor
+from parsl.providers import LocalProvider
+from parsl.usage_tracking.levels import LEVEL_1
+from parsl.concurrent import ParslPoolExecutor
+from academy.logging import init_logging
 from academy.exchange import LocalExchangeFactory
 from academy.manager import Manager
-from academy.logging import init_logging
 
 from cascade.agents.agents import (
     DatabaseMonitor,
-    DynamicsEngine,
+    DynamicsRunner,
     Auditor,
     Sampler,
-    DummyLabeler,
-    DummyTrainer
+    Labeler,
+    Trainer
 )
 from cascade.agents.config import (
     DatabaseConfig,
     DatabaseMonitorConfig,
-    DynamicsEngineConfig,
+    DynamicsRunnerConfig,
     AuditorConfig,
     SamplerConfig,
     LabelerConfig,
@@ -41,7 +43,16 @@
 from cascade.model import AdvanceSpec
 from cascade.learning.mace import MACEInterface
 from cascade.agents.db_orm import TrajectoryDB
-from cascade.agents.task import random_audit, advance_dynamics, random_sample
+from cascade.agents.task import (
+    random_audit,
+    advance_dynamics,
+    random_sample,
+    label_noop,
+    training_noop
+)
+
+# Suppress FutureWarning about torch.load weights_only parameter from MACE
+warnings.filterwarnings("ignore", category=FutureWarning, module="mace.calculators")
 
 
 def parse_args() -> argparse.Namespace:
@@ -139,12 +150,14 @@ def parse_args() -> argparse.Namespace:
 
     return args
 
+
 def get_learner(learner_name: str) -> type[ase.calculators.calculator.Calculator]:
     if learner_name == 'mace':
         return MACEInterface()
     else:
         raise ValueError(f'Unknown learner: {learner_name}')
 
+
 def get_dynamics_cls(cls_name: str) -> type[ase.md.md.MolecularDynamics]:
     if cls_name == 'velocity-verlet':
         return VelocityVerlet
@@ -153,7 +166,6 @@ def get_dynamics_cls(cls_name: str) -> type[ase.md.md.MolecularDynamics]:
 
 
 async def main():
-    
     # parse arguments
     args = parse_args()
 
@@ -172,9 +184,17 @@ async def main():
     logfile = run_dir / "runtime.log"
 
     logger = init_logging(level=args.log_level, logfile=logfile)
+
+    logger.setLevel(logging.DEBUG)
+
     logger.info("Loaded run params")
-    logger.info("Created run directory: %s", run_dir)
-    
+    logger.info(f'Running job in {run_dir}')
+    parsl_logger = logging.getLogger('parsl')
+
+    for handler in parsl_logger.handlers[:]:  # Iterate over a copy of the list
+        parsl_logger.removeHandler(handler)
+    parsl_logger.addHandler(logging.FileHandler(run_dir / 'parsl.log'))
+
     init_strc = args.initial_structures
     learner = get_learner(args.learner)
     init_weights = learner.serialize_model(learner.get_model(mace_mp('small').models[0]))
@@ -209,133 +229,167 @@ async def main():
             )
         )
 
-    # initialize manager, exchange
-    async with await Manager.from_exchange_factory(
-        factory=LocalExchangeFactory(),
-        executors=ThreadPoolExecutor(max_workers=10),
-    ) as manager:
-
-        # register all agents with manager
-        db_reg = await manager.register_agent(DatabaseMonitor)
-        trainer_reg = await manager.register_agent(DummyTrainer)
-        labeler_reg = await manager.register_agent(DummyLabeler)
-        sampler_reg = await manager.register_agent(Sampler)
-        auditor_reg = await manager.register_agent(Auditor)
-        dynamics_reg = await manager.register_agent(DynamicsEngine)
-
-        # get handles to all agents
-        db_handle = manager.get_handle(db_reg)
-        trainer_handle = manager.get_handle(trainer_reg)
-        labeler_handle = manager.get_handle(labeler_reg)
-        sampler_handle = manager.get_handle(sampler_reg)
-        auditor_handle = manager.get_handle(auditor_reg)
-        dynamics_handle = manager.get_handle(dynamics_reg)
-
-        handles = [
-            db_handle,
-            trainer_handle,
-            labeler_handle,
-            sampler_handle,
-            auditor_handle,
-            dynamics_handle
-        ]
-
-        # create config objects
-        db_config = DatabaseMonitorConfig(
-            run_id=run_id,
-            db_url=args.db_url,
-            retrain_len=args.retrain_len,
-            target_length=args.target_length,
-            chunk_size=args.chunk_size,
-            retrain_fraction=args.retrain_fraction,
-            retrain_min_frames=args.retrain_min_frames
-        )
-        dynamics_config = DynamicsEngineConfig(
-            run_id=run_id,
-            db_url=args.db_url,
-            #init_specs=initial_specs,
-            learner=learner,
-            weights=init_weights,
-            dyn_cls=get_dynamics_cls(args.dyn_cls),
-            dyn_kws={'timestep': args.dt_fs * units.fs, 'loginterval': args.loginterval},
-            run_kws={}
-        )
-        auditor_config = AuditorConfig(
-            run_id=run_id,
-            db_url=args.db_url,
-            accept_rate=args.accept_rate,
-            chunk_size=args.chunk_size
-        )
-        sampler_config = SamplerConfig(
-            run_id=run_id,
-            db_url=args.db_url,
-            n_frames=args.n_sample_frames
-        )
-        labeler_config = LabelerConfig(run_id=run_id, db_url=args.db_url)
-        trainer_config = TrainerConfig(run_id=run_id, db_url=args.db_url, learner=learner)
-
-        # launch all agents
-        await manager.launch(
-            DatabaseMonitor,
-            args=(
-                db_config,
+    # set up parsl pool
+    # a chunk can only be in one worker at a time + training happens concurrently
+    n_workers = len(initial_specs)+1
+    config = Config(
+        executors=[
+            HighThroughputExecutor(
+                label="htex_local",
+                max_workers_per_node=n_workers,
+                provider=LocalProvider(
+                    init_blocks=1,
+                    max_blocks=1,
+                ),
+            )
+        ],
+        usage_tracking=LEVEL_1,
+    )
+
+    with ParslPoolExecutor(config=config) as pool:
+        async with await Manager.from_exchange_factory(
+            factory=LocalExchangeFactory(),
+            executors=ThreadPoolExecutor(max_workers=5+len(initial_specs)),
+        ) as manager:
+
+            # register all agents with manager
+            db_reg = await manager.register_agent(DatabaseMonitor)
+            trainer_reg = await manager.register_agent(Trainer)
+            labeler_reg = await manager.register_agent(Labeler)
+            sampler_reg = await manager.register_agent(Sampler)
+            auditor_reg = await manager.register_agent(Auditor)
+
+            # get handles to all agents
+            db_handle = manager.get_handle(db_reg)
+            trainer_handle = manager.get_handle(trainer_reg)
+            labeler_handle = manager.get_handle(labeler_reg)
+            sampler_handle = manager.get_handle(sampler_reg)
+            auditor_handle = manager.get_handle(auditor_reg)
+
+            handles = [
+                db_handle,
                 trainer_handle,
-                dynamics_handle,
-            ),
-            registration=db_reg
-        )
-        await manager.launch(
-            DynamicsEngine,
-            args=(
-                dynamics_config,
-                auditor_handle,
-                ProcessPoolExecutor(max_workers=10),
-                advance_dynamics,
-            ),
-            registration=dynamics_reg
-        )
-        await manager.launch(
-            Auditor,
-            args=(
-                auditor_config,
+                labeler_handle,
                 sampler_handle,
-                dynamics_handle,
-                random_audit,
-                ProcessPoolExecutor(max_workers=10)
+                auditor_handle,
+            ]
 
-            ),
-            registration=auditor_reg,
-        )
-        await manager.launch(
-            Sampler,
-            args=(
-                sampler_config,
-                labeler_handle,
-                ProcessPoolExecutor(max_workers=10),
-                random_sample,
-            ),
-            registration=sampler_reg,
-        )
-        await manager.launch(
-            DummyLabeler,
-            args=(labeler_config,),
-            registration=labeler_reg,
-        )
-        await manager.launch(
-            DummyTrainer,
-            args=(trainer_config,),
-            registration=trainer_reg
-        )        
-
-        # submit initial specs to dynamics engine
-        for spec in initial_specs:
-            logger.info(f"Submitting initial spec to dynamics engine: {spec}")
-            await dynamics_handle.submit(spec)
-        try:
-            await manager.wait([db_handle])
-        except KeyboardInterrupt:
-            for handle in handles:
-                await manager.shutdown(handle, blocking=False)
+            db_monitor_config = DatabaseMonitorConfig(
+                run_id=run_id,
+                db_url=args.db_url,
+                retrain_len=args.retrain_len,
+                chunk_size=args.chunk_size,
+                retrain_fraction=args.retrain_fraction,
+                retrain_min_frames=args.retrain_min_frames,
+            )
+            auditor_config = AuditorConfig(
+                audit_task=random_audit,
+                executor=pool,
+                run_id=run_id,
+                db_url=args.db_url,
+                audit_kwargs=dict(accept_prob=args.accept_rate,),
+                chunk_size=args.chunk_size,
+            )
+            sampler_config = SamplerConfig(
+                run_id=run_id,
+                db_url=args.db_url,
+                n_frames=args.n_sample_frames,
+                executor=pool,
+                sample_task=random_sample,
+            )
+            labeler_config = LabelerConfig(
+                run_id=run_id,
+                db_url=args.db_url,
+                executor=pool,
+                label_task=label_noop
+                )
+            trainer_config = TrainerConfig(
+                run_id=run_id,
+                db_url=args.db_url,
+                executor=pool,
+                training_task=training_noop,
+                training_args=(),
+                training_kwargs={},
+                learner=learner
+            )
+
+            # launch all agents
+            await manager.launch(
+                Auditor,
+                kwargs=dict(
+                    config=auditor_config,
+                    sampler=sampler_handle,
+                ),
+                registration=auditor_reg,
+            )
+            await manager.launch(
+                Sampler,
+                kwargs=dict(
+                    config=sampler_config,
+                    labeler=labeler_handle,
+                ),
+                registration=sampler_reg,
+            )
+            await manager.launch(
+                Labeler,
+                kwargs=dict(config=labeler_config),
+                registration=labeler_reg,
+            )
+            await manager.launch(
+                Trainer,
+                kwargs=dict(config=trainer_config),
+                registration=trainer_reg
+            )
+
+            dyn_handles = []
+            for spec in initial_specs:
+                reg = await manager.register_agent(DynamicsRunner)
+                handle = manager.get_handle(reg)
+
+                handles.append(handle)
+                dyn_handles.append(handle)
+                dyn_config = DynamicsRunnerConfig(
+                        atoms=spec.atoms,
+                        run_id=run_id,
+                        db_url=args.db_url,
+                        traj_id=spec.traj_id,
+                        chunk_size=args.chunk_size,
+                        n_steps=args.target_length,
+                        executor=pool,
+                        advance_dynamics_task=advance_dynamics,
+                        learner=learner,
+                        run_dir=run_dir,
+                        weights=init_weights,
+                        dyn_cls=VelocityVerlet,
+                        dyn_kws={'timestep': 1 * units.fs},
+                        run_kws={},
+                        device='cpu',
+                        model_version=0
+                )
+                await manager.launch(
+                    DynamicsRunner,
+                    kwargs=dict(
+                        config=dyn_config,
+                        auditor=auditor_handle
+                    ),
+                    registration=reg
+                )
+
+            await manager.launch(
+                DatabaseMonitor,
+                kwargs=dict(
+                    config=db_monitor_config,
+                    trainer=trainer_handle,
+                    dynamics_runners=dyn_handles
+                ),
+                registration=db_reg,
+            )
+            # wait for it to finish!
+            try:
+                await manager.wait([db_handle])
+            except KeyboardInterrupt:
+                for handle in handles:
+                    await manager.shutdown(handle, blocking=False)
 
 if __name__ == '__main__':
     asyncio.run(main())
diff --git a/3_cascade/run_cascade_academy.sh b/3_cascade/run_cascade_academy.sh
index fa956ee..d272032 100644
--- a/3_cascade/run_cascade_academy.sh
+++ b/3_cascade/run_cascade_academy.sh
@@ -4,14 +4,29 @@ python run_cascade_academy.py \
     --initial-structures \
         ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
         ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
-    --chunk-size 5 \
-    --target-length 10 \
-    --retrain-len 15 \
-    --retrain-fraction 0.75 \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+        ../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp \
+    --chunk-size 10 \
+    --target-length 40 \
+    --retrain-len 80 \
+    --retrain-fraction 0.5 \
     --n-sample-frames 5 \
     --accept-rate .5 \
     --learner mace \
     --calc mace \
     --dyn-cls velocity-verlet \
     --dt_fs 1.0 \
-    --loginterval 1
+    --loginterval 1 \
+    --log-level DEBUG
diff --git a/cascade/agents/agents.py b/cascade/agents/agents.py
index f0448a0..acafa35 100644
--- a/cascade/agents/agents.py
+++ b/cascade/agents/agents.py
@@ -10,476 +10,233 @@
 from __future__ import annotations
 
 import asyncio
-from asyncio import Queue
-from threading import Event
-import gc
+from asyncio import Queue, Event, Lock, wrap_future
+from typing import TYPE_CHECKING
 import logging
-from functools import cached_property
-from typing import Any, Awaitable, Callable, NamedTuple, Optional, cast
-from concurrent.futures import Executor, Future as ConcurrentFuture
+from concurrent.futures import Executor
 
-import os
-import numpy as np
-from ase import Atoms
-from academy.handle import Handle
-from academy.agent import Agent, action, loop
 from ase.optimize.optimize import Dynamics
 from mace.calculators import mace_mp
-from parsl.concurrent import ParslPoolExecutor
-from parsl.config import Config
-
-try:
-    import psutil
-    PSUTIL_AVAILABLE = True
-    _PSUTIL_WARNING_LOGGED = False
-except ImportError:
-    PSUTIL_AVAILABLE = False
-    _PSUTIL_WARNING_LOGGED = False
+from academy.handle import Handle
+from academy.agent import Agent, action, loop
+from academy.exception import AgentTerminatedError
 
 from cascade.learning.base import BaseLearnableForcefield
-from cascade.utils import canonicalize
 from cascade.model import AuditStatus, AdvanceSpec, AuditResult, TrajectoryStatus, ChunkEventType
 from cascade.agents.config import (
-    CascadeAgentConfig,
-    DatabaseConfig,
-    DynamicsEngineConfig,
     AuditorConfig,
     SamplerConfig,
     LabelerConfig,
-    TrainerConfig,
-    DatabaseMonitorConfig
+    DatabaseMonitorConfig,
+    DynamicsRunnerConfig
 )
 from cascade.agents.db_orm import TrajectoryDB
-from cascade.model import ChunkSpec, TrainingFrame, TrainingFrameSpec
-
-
-
-
-
-def count_file_descriptors_by_type() -> dict:
-    """Count open file descriptors by type for the current process
-    
-    Returns:
-        Dict with counts by FD type (REG, PIPE, IPv4, IPv6, etc.)
-    """
-    import os
-    import subprocess
-    
-    pid = os.getpid()
-    type_counts = {}
-    
-    # First, get total count from /proc (fast and reliable)
-    try:
-        fd_dir = f'/proc/{pid}/fd'
-        if os.path.exists(fd_dir):
-            total_fds = len(os.listdir(fd_dir))
-            type_counts['total'] = total_fds
-    except (OSError, PermissionError):
-        pass
-    
-    # Try to get detailed breakdown from lsof, but don't block if it's slow
-    try:
-        result = subprocess.run(
-            ['lsof', '-p', str(pid), '-F', 't'],
-            capture_output=True,
-            text=True,
-            timeout=1  # Reduced timeout to 1 second
-        )
-        
-        if result.returncode == 0:
-            # Count by type (lines starting with 't' are type indicators)
-            for line in result.stdout.split('\n'):
-                if line.startswith('t'):
-                    fd_type = line[1:]  # Remove 't' prefix
-                    type_counts[fd_type] = type_counts.get(fd_type, 0) + 1
-    except (FileNotFoundError, subprocess.TimeoutExpired):
-        # If lsof fails or times out, we still have the total count
-        if 'total' not in type_counts:
-            return {'error': 'lsof unavailable and /proc access failed'}
-    except Exception as e:
-        # Other errors - still return what we have
-        if 'total' not in type_counts:
-            return {'error': f'Exception counting FDs: {e}'}
-    
-    # Return what we have (at minimum, the total count)
-    return type_counts if type_counts else {'error': 'Could not count FDs'}
-
-
-def check_executor_workers(executor: Executor) -> dict:
-    """Check health of ProcessPoolExecutor worker processes
-    
-    Args:
-        executor: ProcessPoolExecutor instance (or any Executor)
-        
-    Returns:
-        Dict with worker process health info, or error dict if unavailable
-    """
-    global _PSUTIL_WARNING_LOGGED
-    if not PSUTIL_AVAILABLE:
-        if not _PSUTIL_WARNING_LOGGED:
-            logger = logging.getLogger(__name__)
-            logger.warning(
-                "psutil not available - cannot monitor executor worker health. "
-                "Install psutil to enable worker process monitoring."
-            )
-            _PSUTIL_WARNING_LOGGED = True
-        return {'error': 'psutil not available'}
-    
-    # Only check ProcessPoolExecutor, skip others
-    from concurrent.futures import ProcessPoolExecutor
-    if not isinstance(executor, ProcessPoolExecutor):
-        return {'skipped': 'Not a ProcessPoolExecutor'}
-    
-    try:
-        if not hasattr(executor, '_processes'):
-            return {'error': 'Cannot access executor processes'}
-        
-        alive = 0
-        dead = 0
-        zombies = 0
-        worker_pids = []
-        
-        for proc in executor._processes.values():
-            if proc is None:
-                continue
-            
-            try:
-                pid = proc.pid
-                worker_pids.append(pid)
-                p = psutil.Process(pid)
-                status = p.status()
-                
-                if status == psutil.STATUS_ZOMBIE:
-                    zombies += 1
-                elif p.is_running():
-                    alive += 1
-                else:
-                    dead += 1
-            except (psutil.NoSuchProcess, psutil.AccessDenied):
-                dead += 1
-            except Exception as e:
-                # Catch any other unexpected exceptions from psutil
-                logging.getLogger(__name__).debug(f"Error checking process {proc.pid if hasattr(proc, 'pid') else 'unknown'}: {e}")
-                dead += 1
-        
-        return {
-            'alive': alive,
-            'dead': dead,
-            'zombies': zombies,
-            'worker_pids': worker_pids,
-            'total_workers': len(executor._processes)
-        }
-    except Exception as e:
-        # Catch any unexpected exceptions (e.g., AttributeError accessing _processes)
-        logging.getLogger(__name__).debug(f"Error in check_executor_workers: {e}")
-        return {'error': f'Exception checking workers: {e}'}
+from cascade.model import ChunkSpec, TrainingFrameSpec, Chunk
+
+if TYPE_CHECKING:
+    from typing import Callable
+    from ase import Atoms
 
 
 class CascadeAgent(Agent):
     """Base class for all cascade agents"""
-    
-    # Track resource creation across all agents
-    _resource_counts = {
-        'agents_created': 0,
-        'traj_db_instances_created': 0,
-    }
 
-    def __init__(
-        self,
-        config: CascadeAgentConfig,
-        executor: Optional[Executor] = None
-    ):
-        self.config = config
+    async def agent_on_startup(self):
         self.logger = logging.getLogger(self.__class__.__name__)
-        self.executor = executor
-        CascadeAgent._resource_counts['agents_created'] += 1
-        self.logger.debug(
-            f"Agent created. Total agents: {CascadeAgent._resource_counts['agents_created']}"
-        )
-    
-    def schedule_future_callback(
-        self,
-        future: ConcurrentFuture[Any],
-        callback: Callable[[ConcurrentFuture[Any]], Awaitable[None]],
-    ) -> None:
-        """Schedule a callback to run after a future completes.
+        self._traj_db = TrajectoryDB(self.db_url, logger=self.logger)
+        self._traj_db.create_tables()
 
-        Args:
-            future: Future from executor.submit() (concurrent.futures.Future)
-            callback: Coroutine accepting the completed future.
-        """
-        async def _run_callback() -> None:
-            await asyncio.wrap_future(future)  # make concurrent future awatable 
-            await callback(future)
-
-        asyncio.create_task(_run_callback())
-    
-    @cached_property
-    def _traj_db(self) -> TrajectoryDB:
-        """A TrajectoryDB object for managing trajectory and chunk metadata"""
-        CascadeAgent._resource_counts['traj_db_instances_created'] += 1
-        self.logger.info(
-            f"Creating TrajectoryDB instance #{CascadeAgent._resource_counts['traj_db_instances_created']} "
-            f"for {self.__class__.__name__}"
-        )
-        traj_db = TrajectoryDB(self.config.db_url, logger=self.logger)
-        traj_db.create_tables()  # Ensure tables exist
-        return traj_db
 
-    @loop
-    async def monitor_executor_health(self, shutdown: asyncio.Event) -> None:
-        """Periodically check executor worker process health"""
-        self.logger.info(f"Monitoring executor health for {self.config.run_id}")
-        while not shutdown.is_set():
-            try:
-                # Check if executor exists
-                if self.executor is None:
-                    self.logger.debug("Executor is None, skipping health check")
-                    await asyncio.sleep(30)
-                    continue
-                
-                # Get worker health status
-                worker_health = check_executor_workers(self.executor)
-                self.logger.info(f"check result: {worker_health}")
-                
-                # Handle errors or skipped checks
-                if 'error' in worker_health:
-                    self.logger.debug(f"Executor health check error: {worker_health['error']}")
-                    await asyncio.sleep(30)
-                    continue
-                
-                if 'skipped' in worker_health:
-                    self.logger.debug(f"Executor health check skipped: {worker_health['skipped']}")
-                    await asyncio.sleep(30)
-                    continue
-                
-                # Check for zombie processes
-                zombies = worker_health.get('zombies', 0)
-                if zombies > 0:
-                    self.logger.warning(
-                        f"Found {zombies} zombie worker processes: {worker_health}"
-                    )
-                    await asyncio.sleep(30)
-                    continue
-                
-                # Check for many dead workers
-                dead = worker_health.get('dead', 0)
-                alive = worker_health.get('alive', 0)
-                if dead > alive:
-                    self.logger.warning(
-                        f"Many dead workers detected: {worker_health}. Consider recreating executor."
-                    )
-                    await asyncio.sleep(30)
-                    continue
-                
-                # All good - log normal health status
-                self.logger.info(f"Executor worker health: {worker_health}")
-                
-            except Exception as e:
-                self.logger.exception(f"Error in executor health monitoring: {e}")
-            
-            await asyncio.sleep(30)  # Check every 30 seconds
-
-class DynamicsEngine(CascadeAgent):
+class DynamicsRunner(CascadeAgent):
 
     def __init__(
         self,
-        config: DynamicsEngineConfig,
         auditor: Handle[Auditor],
-        executor: Executor,
-        advance_dynamics_task: Callable[[AdvanceSpec], None]
+        config: DynamicsRunnerConfig
     ):
-        super().__init__(config, executor)
-        self.weights = config.weights  # This needs to be mutable
-        self.model_version = 0  # todo: mt.2025.10.20 probably this should be persisted somewhere else
-        self.queue = Queue()
+        """Runs dynamics in a loop until done, or a shutdown message is received
+
+        Arguments:
+            atoms: initial conditions
+            run_id: which run this is (for logging) # todo surely we dont have to pass this to every agent like this
+            chunk_size: how many steps to advance at a time
+            n_steps: total number of timesteps to run
+            auditor: handle to auditor class
+            executor: where the dynamics is executed
+            advance_dynamics_task: task run on the executor
+            learner: cascade learner class
+            weights: weights for the learner
+            dyn_cls: ase dynamics class
+            dyn_kws: arguments to the dynamics constructor
+            run_kws: arguments to the dynamics run method
+            device: for torch execution
+            model_version: index of current model version
+        """
+        self.db_url = config.db_url
+        super().__init__()
+        self.config = config
         self.auditor = auditor
-        self.advance_dynamics_task = advance_dynamics_task
 
-    # async def agent_on_startup(self):
-    #     for spec in self.config.init_specs:
-    #         await self.queue.put(spec)
+        # pull out variables that may change from config
+        self.atoms = config.atoms.copy()
+        self.init_chunk_size = config.chunk_size
+        self.chunk_size = config.chunk_size
+        self.model_version = config.model_version
+        self.weights = config.weights
 
-    @action
-    async def receive_weights(self, weights: bytes) -> None:
-        self.weights = weights
-        self.model_version += 1
-        self.logger.info(f"Received new weights, now on model version {self.model_version}")
+        # track progress
+        self.timestep = 0
+        self.chunk_ix = 0
+        self.attempt = 0
+        self.done = False
 
-    @action
-    async def submit(self, spec: AdvanceSpec):
-        if isinstance(spec, dict):
-            spec = AdvanceSpec(**spec)
-        self._traj_db.mark_trajectory_running(self.config.run_id, spec.traj_id)
-        self._traj_db.record_chunk_event(
-            run_id=self.config.run_id,
-            traj_id=spec.traj_id,
-            chunk_id=spec.chunk_id,
-            attempt_index=spec.attempt_index,
-            event_type=ChunkEventType.STARTED_DYNAMICS
-        )
-        self.logger.info(f"Received advance spec for traj {spec.traj_id} chunk {spec.chunk_id} attempt {spec.attempt_index}")
-
-        initial_future = self.executor.submit(
-            self.advance_dynamics_task,
-            spec=spec,
-            learner=self.config.learner,
-            weights=self.weights,
-            db_url=self.config.db_url,
-            device=self.config.device,
-            dyn_cls=self.config.dyn_cls,
-            dyn_kws=self.config.dyn_kws
-        )
+        # for handling weight updates
+        self.received_weights = Event()
+        self.new_model_lock = Lock()
+        self.new_model: tuple[bytes, int] | None = None  # weights, version
 
-        self.schedule_future_callback(
-            initial_future,
-            self._advance_dynamics_callback,
-        )
-    async def _advance_dynamics_callback(
-        self,
-        future: ConcurrentFuture[Any]
+    @loop
+    async def run(
+            self,
+            shutdown: asyncio.Event,
     ) -> None:
-        """Handle the results of an advance dynamics call
-        
-        Args:
-            future: The future that contains the result of the advance dynamics call
+        """Run dynamics until done, or a shutdown message is received"""
+        while not (shutdown.is_set() or self.done):
+
+            # there are two conditions to release this lock: we finish, or we are waiting for new weights
+            # await self.weights_lock.acquire()
+            # submit dynamics for evaluation
+            spec = AdvanceSpec(
+                atoms=self.atoms,
+                steps=self.chunk_size,
+                run_id=self.config.run_id,
+                traj_id=self.config.traj_id,
+                chunk_id=self.chunk_ix,
+                attempt_index=self.attempt,
+            )
 
-        Returns:
-            None
+            self.logger.info(f"Running dynamics for traj {spec.traj_id} chunk {spec.chunk_id} attempt {spec.attempt_index} with {spec.steps} steps")
+
+            async with self.new_model_lock:
+
+                if self.new_model:
+                    self.weights, self.model_version = self.new_model
+                    self.new_model = None
+                self.logger.debug(
+                    f"Submitting dynamics to executor dynamics for traj {spec.traj_id} chunk {spec.chunk_id} attempt {spec.attempt_index} with {spec.steps} steps")
+                chunk_future = self.config.executor.submit(
+                    self.config.advance_dynamics_task,
+                    spec=spec,
+                    learner=self.config.learner,
+                    weights=self.config.weights,
+                    db_url=self.config.db_url,
+                    device=self.config.device,
+                    dyn_cls=self.config.dyn_cls,
+                    dyn_kws=self.config.dyn_kws,
+                    run_kws=self.config.run_kws,
+                    run_dir=str(self.config.run_dir)
+                )
 
-        Track the chunk metadata in the database and submit to auditor.
-        """
-        self.logger.debug('Advance dynamics callback started')
-        self.logger.info('Getting future result')
-        spec = future.result()
-        if isinstance(spec, dict):
-            spec = AdvanceSpec(**spec)
-
-        run_id = spec.run_id
-        traj_id = spec.traj_id
-        chunk_id = spec.chunk_id
-        attempt_index = spec.attempt_index
-        steps = spec.steps
-
-        # Calculate number of frames from steps and loginterval
-        loginterval = self.config.dyn_kws.get('loginterval', 1)
-        n_frames = steps // loginterval
-
-        # Record chunk metadata in ORM
-        success = self._traj_db.add_chunk_attempt(
-            run_id=run_id,
-            traj_id=traj_id,
-            chunk_id=chunk_id,
-            model_version=self.model_version,
-            n_frames=n_frames,
-            audit_status=AuditStatus.PENDING,
-            attempt_index=attempt_index
-        )
-        if success:
-            self.logger.info(f"Recorded chunk {chunk_id} of traj {traj_id} in database (attempt {attempt_index}, {n_frames} frames)")
-            self._traj_db.record_chunk_event(
-                run_id=run_id,
-                traj_id=traj_id,
-                chunk_id=chunk_id,
-                attempt_index=attempt_index,
-                event_type=ChunkEventType.FINISHED_DYNAMICS
+            self._traj_db.add_chunk_attempt(
+                run_id=self.config.run_id,
+                traj_id=spec.traj_id,
+                chunk_id=spec.chunk_id,
+                model_version=self.model_version,
+                n_frames=self.chunk_size,
+                audit_status=AuditStatus.PENDING,
+                attempt_index=self.attempt
             )
-        else:
-            self.logger.error(f"Failed to record chunk {chunk_id} of traj {traj_id} in database (attempt {attempt_index})")
 
-        # submit to auditor
-        self.logger.info(f"Submitting audit for chunk {chunk_id} of traj {traj_id} attempt {attempt_index}.")
-        await self.auditor.submit(ChunkSpec(traj_id=traj_id, chunk_id=chunk_id))
+            # get future result
+            wrapped_future = wrap_future(chunk_future)
+            await wrapped_future
+            chunk_atoms = wrapped_future.result()
+
+            # write atoms # todo wrap this up
+            for frame_index, _atoms in enumerate(chunk_atoms):
+                frame_index += self.timestep
+                self._traj_db.write_frame(
+                    run_id=spec.run_id,
+                    traj_id=spec.traj_id,
+                    chunk_id=spec.chunk_id,
+                    attempt_index=spec.attempt_index,
+                    frame_index=frame_index,
+                    atoms=_atoms
+                )
+            self.logger.info(f"Finished dynamics for traj {spec.traj_id} chunk {spec.chunk_id} attempt {spec.attempt_index}")
+
+            # submit to auditor
+            chunk = Chunk(
+                atoms=chunk_atoms,
+                traj_id=self.config.traj_id,
+                chunk_id=self.chunk_ix,
+                attempt_ix=self.attempt_ix,
+                model_version=self.model_version
+            )
+            self.logger.info(f"Submitting audit for traj {self.config.traj_id} chunk {spec.chunk_id} attempt {spec.attempt_index}")
+            audit_result = await self.auditor.audit(chunk)
+
+            # handle audit result
+            if audit_result.status == AuditStatus.PASSED:
+
+                self.logger.info(f"Audit status passed for traj {self.config.traj_id} chunk {self.chunk_ix} attempt {self.attempt}")
+                self.timestep += self.chunk_size
+                self.logger.info(f"On timestep {self.timestep} of {self.config.n_steps}")
+                self.done = self.timestep >= self.config.n_steps
+                if self.done:
+                    self.logger.info(f"Finished dynamics for traj {self.config.traj_id} on {self.chunk_ix} attempt {self.attempt}, shutting down")
+                    self._traj_db.mark_trajectory_completed(run_id=self.config.run_id, traj_id=self.config.traj_id)
+                    self.agent_shutdown()
+                else:
+                    # audit passed but not done, use the new atoms to run a new chunk
+                    self.atoms = chunk_atoms[-1]
+                    self.chunk_ix += 1
+                    self.attempt = 0
+                    self.logger.info(f"Updating traj {self.config.traj_id} to chunk {self.chunk_ix} attempt {self.attempt}")
+            else:
+                # audit failed
+                self.logger.info(f'Audit status failed for traj {self.config.traj_id} chunk {self.chunk_ix} attempt {self.attempt}, waiting for new weights...')
+                self.attempt += 1
+                self.received_weights.clear()
+                await self.received_weights.wait()
+                self.logger.info('Received new weights')
+
+    @action
+    async def receive_weights(self, weights: bytes, model_version: int) -> None:
+        async with self.new_model_lock:
+            self.new_model = (weights, model_version)
+        self.logger.info(f"Received weights for model version {model_version}")
+        self.received_weights.set()
 
 
 class Auditor(CascadeAgent):
 
     def __init__(
             self,
-            config: AuditorConfig,
             sampler: Handle[Sampler],
-            dynamics_engine: Handle[DynamicsEngine],
-            audit_task: Callable[[ChunkSpec], AuditResult],
-            executor: Executor
+            config=AuditorConfig
     ):
-        super().__init__(config, executor)
+        self.db_url = config.db_url
+        super().__init__()
+        self.config = config
         self.sampler = sampler
-        self.dynamics_engine = dynamics_engine
         self.queue = Queue()
-        self.audit_task = audit_task
+        self.chunk_size = config.chunk_size
 
     @action
-    async def submit(self, chunk_spec: ChunkSpec):
+    async def audit(self, chunk: Chunk) -> AuditResult:
         """Submit a chunk for audit"""
-        self.logger.info(f'Received chunk {chunk_spec.chunk_id} from traj {chunk_spec.traj_id}')
+        self.logger.info(f'Submitting audit of traj {chunk.traj_id} chunk {chunk.chunk_id} attempt {chunk.attempt_ix} to executor')
 
-        latest_attempt = self._traj_db.get_latest_chunk_attempt(
-            run_id=self.config.run_id,
-            traj_id=chunk_spec.traj_id,
-            chunk_id=chunk_spec.chunk_id
-        )
-        if not latest_attempt:
-            self.logger.warning(
-                'No attempt metadata found for traj %s chunk %s; skipping audit',
-                chunk_spec.traj_id,
-                chunk_spec.chunk_id,
-            )
-            return
-        attempt_index = latest_attempt['attempt_index']
-        self._traj_db.record_chunk_event(
-            run_id=self.config.run_id,
-            traj_id=chunk_spec.traj_id,
-            chunk_id=chunk_spec.chunk_id,
-            attempt_index=attempt_index,
-            event_type=ChunkEventType.STARTED_AUDIT
-        )
-        chunk_atoms = self._traj_db.get_latest_chunk_attempt_atoms(
-            self.config.run_id,
-            chunk_spec.traj_id,
-            chunk_spec.chunk_id
+        future = self.config.executor.submit(
+            self.config.audit_task,
+            chunk,
+            **self.config.audit_kwargs
         )
-
-        self.logger.info(f'Submitting audit of chunk {chunk_spec.chunk_id} of traj {chunk_spec.traj_id} to executor')
-
-        initial_future = self.executor.submit(
-            self.audit_task,
-            chunk_atoms=chunk_atoms,
-            chunk_spec=chunk_spec,
-            attempt_index=latest_attempt['attempt_index'],
-        )
-
-        self.schedule_future_callback(
-            initial_future,
-            self._audit_callback,
-        )
-    async def _audit_callback(self, future: ConcurrentFuture[Any]) -> None:
-        """Handle the results of an audit
-    
-        If the audit fails, the chunk is submitted to the sampler.
-        If the audit passes and the trajectory is done, this is recorded in the database.
-        If the audit passes and the trajectory is NOT done, the next chunk is submitted to the dynamics engine.
-
-        Args:
-            future: The future that contains the result of the audit
-
-        Returns:
-            None
-        """
-        self.logger.debug('Audit callback started')
-        self.logger.info('Getting future result')
-        result = future.result()
+        wrapped_future = wrap_future(future)
+        await wrapped_future
+        result = wrapped_future.result()
         status = result.status
-        self.logger.info(
-            'Audit result status: %s (traj_id=%d, chunk_id=%d, attempt_index=%d)',
-            status, result.traj_id, result.chunk_id, result.attempt_index
-        )
-
-        if status not in [AuditStatus.PASSED, AuditStatus.FAILED]:
-            self.logger.error(
-                'Audit result is not PASSED or FAILED: %s (traj_id=%d, chunk_id=%d, attempt_index=%d)',
-                status, result.traj_id, result.chunk_id, result.attempt_index
-            )
-            return
 
         self._traj_db.update_chunk_audit_done_status(
             run_id=self.config.run_id,
@@ -488,325 +245,210 @@ async def _audit_callback(self, future: ConcurrentFuture[Any]) -> None:
             attempt_index=result.attempt_index,
             audit_status=status
         )
-        
-        # Record audit result event
-        event_type = ChunkEventType.AUDIT_PASSED if status == AuditStatus.PASSED else ChunkEventType.AUDIT_FAILED
-        self._traj_db.record_chunk_event(
-            run_id=self.config.run_id,
-            traj_id=result.traj_id,
-            chunk_id=result.chunk_id,
-            attempt_index=result.attempt_index,
-            event_type=event_type
-        )
-        
         if status == AuditStatus.PASSED:
             self.logger.info(
                 f'Audit passed for traj {result.traj_id} chunk {result.chunk_id} attempt {result.attempt_index}'
             )
-            
-            # Check if trajectory is done using the data model
-            done = self._traj_db.is_trajectory_done(
-                run_id=self.config.run_id,
-                traj_id=result.traj_id
-            )
-            if done:
-                self.logger.info(f"Traj {result.traj_id} is complete")
-                self._traj_db.record_chunk_event(
-                    run_id=self.config.run_id,
-                    traj_id=result.traj_id,
-                    chunk_id=result.chunk_id,
-                    attempt_index=result.attempt_index,
-                    event_type=ChunkEventType.TRAJECTORY_COMPLETED
-                )
-            else:
-                # Trajectory not done - submit next chunk
-                # Get the last frame from the current chunk to use as starting point
-                last_frame = self._traj_db.get_last_frame_from_chunk(
-                    run_id=self.config.run_id,
-                    traj_id=result.traj_id,
-                    chunk_id=result.chunk_id,
-                    attempt_index=result.attempt_index
-                ) # todo maybe we want to just query once
-                # Force garbage collection after retrieving atoms
-                gc.collect()
-                
-                # Create and submit next advance spec
-                next_chunk_id = result.chunk_id + 1
-                next_attempt_index = 0
-                next_spec = AdvanceSpec(
-                    atoms=last_frame,
-                    run_id=self.config.run_id,
-                    traj_id=result.traj_id,
-                    chunk_id=next_chunk_id,
-                    attempt_index=next_attempt_index,
-                    steps=self.config.chunk_size
-                )
-                await self.dynamics_engine.submit(next_spec)
-                self.logger.info(f"Submitted next chunk {result.chunk_id + 1} for traj {result.traj_id} (attempt {next_attempt_index}, after chunk {result.chunk_id} attempt {result.attempt_index} passed)")
         else:
             # audit failed, submit to sampler
             self.logger.info(
                 f'Audit failed for traj {result.traj_id} chunk {result.chunk_id} attempt {result.attempt_index}'
             )
-            spec = ChunkSpec(
-                traj_id=result.traj_id,
-                chunk_id=result.chunk_id,
-                attempt_index=result.attempt_index
-            )
             self.logger.info(f'Submitting failed chunk {result.chunk_id} of traj {result.traj_id} to sampler')
-            await self.sampler.submit(spec)
+            asyncio.create_task(self.sampler.submit(chunk))
+        return status
+
 
-    
 class Sampler(CascadeAgent):
 
     def __init__(
         self,
         config: SamplerConfig,
-        labeler: Handle[DummyLabeler],
-        executor: Executor,
-        sample_task: Callable[..., list[TrainingFrameSpec]],
+        labeler: Handle[Labeler],
     ):
-        super().__init__(config, executor)
+        self.db_url = config.db_url
+        super().__init__()
+        self.config = config
         self.queue = Queue()
         self.labeler = labeler
-        self.sample_task = sample_task
+        self.n_frames = config.n_frames
 
-    @action
-    async def submit(self, chunk_spec: ChunkSpec):
-        await self.queue.put(chunk_spec)
-
-    def _make_sample_callback(self, chunk_spec: ChunkSpec):
-        async def _sample_frames_callback(future: ConcurrentFuture[Any]) -> None:
-            specs = future.result()
-            if len(specs) != self.config.n_frames:
-                self.logger.warning(
-                    "Sampling returned %d frames for traj %s chunk %s (attempt %s), "
-                    "expected n_frames=%d",
-                    len(specs),
-                    chunk_spec.traj_id,
-                    chunk_spec.chunk_id,
-                    chunk_spec.attempt_index,
-                    self.config.n_frames,
-                )
-            for spec in specs:
-                self.logger.info(
-                    f'Submitting training frame from traj {chunk_spec.traj_id} '
-                    f'chunk {chunk_spec.chunk_id} to labeler'
-                )
-                await self.labeler.submit(spec)
-            self._traj_db.record_chunk_event(
-                run_id=self.config.run_id,
-                traj_id=chunk_spec.traj_id,
-                chunk_id=chunk_spec.chunk_id,
-                attempt_index=chunk_spec.attempt_index,
-                event_type=ChunkEventType.FINISHED_SAMPLING,
-            )
-        return _sample_frames_callback
-
-    @loop
     async def sample_frames(
         self,
-        shutdown: asyncio.Event
+        chunk: Chunk,
     ) -> None:
-        while not shutdown.is_set():
-            chunk_spec = await self.queue.get()
 
-            self.logger.info(
-                f'Sampling frames from chunk {chunk_spec.chunk_id} of '
-                f'traj {chunk_spec.traj_id}'
-            )
-
-            # Resolve model_version and attempt_index (auditor submits ChunkSpec
-            # without model_version; we store it in chunk metadata)
-            model_version = chunk_spec.model_version
-            attempt_index = chunk_spec.attempt_index
-            if model_version is None or attempt_index is None:
-                latest = self._traj_db.get_latest_chunk_attempt(
-                    self.config.run_id,
-                    chunk_spec.traj_id,
-                    chunk_spec.chunk_id,
-                )
-                if latest is not None:
-                    if model_version is None:
-                        model_version = latest.get('model_version')
-                    if attempt_index is None:
-                        attempt_index = latest['attempt_index']
-            if model_version is None:
-                self.logger.warning(
-                    "Cannot determine model_version for traj %s chunk %s "
-                    "(attempt %s), skipping",
-                    chunk_spec.traj_id,
-                    chunk_spec.chunk_id,
-                    attempt_index,
-                )
-                continue
-
-            # Ensure we have attempt_index for downstream use
-            resolved_spec = ChunkSpec(
-                traj_id=chunk_spec.traj_id,
-                chunk_id=chunk_spec.chunk_id,
-                attempt_index=attempt_index,
-                model_version=model_version,
-            )
-
-            # Get frames from trajectory_frames table for this chunk
-            atoms_list = self._traj_db.get_latest_chunk_attempt_atoms(
-                run_id=self.config.run_id,
-                traj_id=chunk_spec.traj_id,
-                chunk_id=chunk_spec.chunk_id,
-            )
-
-            if not atoms_list:
-                self.logger.warning(
-                    f"No frames found for chunk {chunk_spec.chunk_id} of "
-                    f"traj {chunk_spec.traj_id}"
-                )
-                continue
-
-            # Get frame IDs for the sampled frames
-            frame_ids = self._traj_db.get_chunk_frame_ids(
-                run_id=self.config.run_id,
-                traj_id=chunk_spec.traj_id,
-                chunk_id=chunk_spec.chunk_id,
-                attempt_index=attempt_index,
-            )
-
-            # Record STARTED_SAMPLING (only when we have atoms to process)
-            self._traj_db.record_chunk_event(
-                run_id=self.config.run_id,
-                traj_id=resolved_spec.traj_id,
-                chunk_id=resolved_spec.chunk_id,
-                attempt_index=resolved_spec.attempt_index,
-                event_type=ChunkEventType.STARTED_SAMPLING,
-            )
+        self.logger.info(
+            f'Sampling frames from traj {chunk.traj_id} '
+            f'chunk {chunk.chunk_id} '
+            f'attempt {chunk.attempt_ix}'
+        )
+        future = self.config.executor.submit(
+            self.config.sample_task,
+            chunk,
+            n_frames=self.config.n_frames,
+        )
+        wrapped_future = wrap_future(future)
+        await wrapped_future
+        training_frames = wrapped_future.result()
 
-            future = self.executor.submit(
-                self.sample_task,
-                atoms_list=atoms_list,
-                frame_ids=frame_ids,
-                chunk_spec=resolved_spec,
-                model_version=model_version,
-                n_frames=self.config.n_frames,
+        if len(training_frames) != self.config.n_frames:
+            self.logger.warning(
+                "Sampling returned %d frames for traj %s chunk %s (attempt %s), "
+                "expected n_frames=%d",
+                len(training_frames),
+                chunk.traj_id,
+                chunk.chunk_id,
+                chunk.attempt_ix,
+                self.n_frames,
             )
-            self.schedule_future_callback(
-                future, self._make_sample_callback(resolved_spec)
+        for frame in training_frames:
+            self.logger.info(
+                f'Submitting training frame from traj {chunk.traj_id} '
+                f'chunk {chunk.chunk_id} attempt {chunk.attempt_ix} to labeler'
             )
+            asyncio.create_task(self.labeler.label_data(frame))
 
 
-class DummyLabeler(CascadeAgent):
+class Labeler(CascadeAgent):
 
     def __init__(
         self,
-        config: LabelerConfig,
+        config: LabelerConfig
     ):
-        super().__init__(config)
+        self.db_url = config.db_url
+        super().__init__()
+        self.config = config
         self.queue = Queue()
 
-    @action
-    async def submit(self, training_frame_spec: TrainingFrameSpec) -> None:
-        self.logger.info(f'Received training frame (trajectory_frame_id={training_frame_spec.trajectory_frame_id})')
-        await self.queue.put(training_frame_spec)
-
     @loop
-    async def label_data(self, shutdown: asyncio.Event) -> None:
+    async def label_data(self, frame: TrainingFrame) -> None:
 
-        while not shutdown.is_set():
-            training_frame_spec = await self.queue.get()
-            
-            # Check if STARTED_LABELING exists for this chunk (idempotent)
-            if not self._traj_db.has_chunk_event(
-                run_id=self.config.run_id,
-                traj_id=training_frame_spec.traj_id,
-                chunk_id=training_frame_spec.chunk_id,
-                attempt_index=training_frame_spec.attempt_index,
-                event_type=ChunkEventType.STARTED_LABELING
-            ):
-                self._traj_db.record_chunk_event(
-                    run_id=self.config.run_id,
-                    traj_id=training_frame_spec.traj_id,
-                    chunk_id=training_frame_spec.chunk_id,
-                    attempt_index=training_frame_spec.attempt_index,
-                    event_type=ChunkEventType.STARTED_LABELING
-                )
-            
-            # Record STARTED_LABELING_FRAME
+        # Check if STARTED_LABELING exists for this chunk (idempotent)
+        if not self._traj_db.has_chunk_event(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index,
+            event_type=ChunkEventType.STARTED_LABELING
+        ):
             self._traj_db.record_chunk_event(
                 run_id=self.config.run_id,
-                traj_id=training_frame_spec.traj_id,
-                chunk_id=training_frame_spec.chunk_id,
-                attempt_index=training_frame_spec.attempt_index,
-                event_type=ChunkEventType.STARTED_LABELING_FRAME,
-                frame_id=training_frame_spec.trajectory_frame_id
+                traj_id=frame.traj_id,
+                chunk_id=frame.chunk_id,
+                attempt_index=frame.attempt_index,
+                event_type=ChunkEventType.STARTED_LABELING
             )
-            
-            try:
-                self._traj_db.add_training_frame(
-                    run_id=self.config.run_id,
-                    trajectory_frame_id=training_frame_spec.trajectory_frame_id,
-                    model_version_sampled_from=training_frame_spec.training_frame.model_version,
-                    traj_id=training_frame_spec.traj_id,
-                    chunk_id=training_frame_spec.chunk_id,
-                    attempt_index=training_frame_spec.attempt_index
-                )
-            except Exception as e:
-                self.logger.warning(
-                    f"Could not add training frame for trajectory_frame_id "
-                    f"{training_frame_spec.trajectory_frame_id}: {e}, skipping training frame"
-                )
-                continue
-            
-            # Record FINISHED_LABELING_FRAME after successfully adding training frame
+
+        # Record STARTED_LABELING_FRAME
+        self._traj_db.record_chunk_event(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index,
+            event_type=ChunkEventType.STARTED_LABELING_FRAME,
+            frame_id=frame.trajectory_frame_id
+        )
+
+        # Check if STARTED_LABELING exists for this chunk (idempotent)
+        if not self._traj_db.has_chunk_event(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index,
+            event_type=ChunkEventType.STARTED_LABELING
+        ):
             self._traj_db.record_chunk_event(
                 run_id=self.config.run_id,
-                traj_id=training_frame_spec.traj_id,
-                chunk_id=training_frame_spec.chunk_id,
-                attempt_index=training_frame_spec.attempt_index,
-                event_type=ChunkEventType.FINISHED_LABELING_FRAME,
-                frame_id=training_frame_spec.trajectory_frame_id
+                traj_id=frame.traj_id,
+                chunk_id=frame.chunk_id,
+                attempt_index=frame.attempt_index,
+                event_type=ChunkEventType.STARTED_LABELING
             )
-            
-            # Check if all frames for chunk are done
-            labeled_count = self._traj_db.count_labeled_frames_for_chunk(
+
+        # Record STARTED_LABELING_FRAME
+        self._traj_db.record_chunk_event(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index,
+            event_type=ChunkEventType.STARTED_LABELING_FRAME,
+            frame_id=frame.trajectory_frame_id
+        )
+
+        frame_future = self.config.executor.submit(
+            self.config.label_task,
+            frame
+        )
+        wrapped_future = wrap_future(frame_future)
+        await wrapped_future
+        frame = wrapped_future.result()
+        self._traj_db.add_training_frame(
+            run_id=self.config.run_id,
+            trajectory_frame_id=frame.trajectory_frame_id,
+            model_version_sampled_from=frame.training_frame.model_version,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index
+        )
+
+        # Record FINISHED_LABELING_FRAME after successfully adding training frame
+        self._traj_db.record_chunk_event(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index,
+            event_type=ChunkEventType.FINISHED_LABELING_FRAME,
+            frame_id=frame.trajectory_frame_id
+        )
+
+        # Check if all frames for chunk are done
+        labeled_count = self._traj_db.count_labeled_frames_for_chunk(
+            run_id=self.config.run_id,
+            traj_id=frame.traj_id,
+            chunk_id=frame.chunk_id,
+            attempt_index=frame.attempt_index
+        )
+
+        if labeled_count >= frame.total_frames_in_chunk:
+            # All frames labeled, record FINISHED_LABELING
+            self._traj_db.record_chunk_event(
                 run_id=self.config.run_id,
-                traj_id=training_frame_spec.traj_id,
-                chunk_id=training_frame_spec.chunk_id,
-                attempt_index=training_frame_spec.attempt_index
-            )
-            
-            if labeled_count >= training_frame_spec.total_frames_in_chunk:
-                # All frames labeled, record FINISHED_LABELING
-                self._traj_db.record_chunk_event(
-                    run_id=self.config.run_id,
-                    traj_id=training_frame_spec.traj_id,
-                    chunk_id=training_frame_spec.chunk_id,
-                    attempt_index=training_frame_spec.attempt_index,
-                    event_type=ChunkEventType.FINISHED_LABELING
-                )
-            
-            self.logger.info(
-                f"Added training frame to database: traj={training_frame_spec.traj_id}, "
-                f"chunk={training_frame_spec.chunk_id}, attempt={training_frame_spec.attempt_index}, "
-                f"model_version={training_frame_spec.training_frame.model_version}"
+                traj_id=frame.traj_id,
+                chunk_id=frame.chunk_id,
+                attempt_index=frame.attempt_index,
+                event_type=ChunkEventType.FINISHED_LABELING
             )
 
+        self.logger.info(
+            f"Added training frame to database: traj={frame.traj_id}, "
+            f"chunk={frame.chunk_id}, attempt={frame.attempt_index}, "
+            f"model_version={frame.training_frame.model_version}"
+        )
+
+
+class Trainer(CascadeAgent):
 
-class DummyTrainer(CascadeAgent):
+    def __init__(self, config):
+        self.db_url = config.db_url
+        super().__init__()
+        self.config = config
 
     @action
     async def train_model(
         self,
-        training_round: int
+        training_round: int,
     ) -> bytes:
-        # Record STARTED_TRAINING event
-        self._traj_db.record_training_event(
-            run_id=self.config.run_id,
-            event_type=ChunkEventType.STARTED_TRAINING,
-            training_round=training_round
+
+        training_future = self.config.executor.submit(
+            self.config.training_task,
+            self.config.learner,
+            *self.config.training_args,
+            **self.config.training_kwargs
         )
-        
-        calc = mace_mp('small', device='cpu', default_dtype="float32") #todo: mt.2025.11.04 this should be configurable
-        model = calc.models[0]
-        model_msg = self.config.learner.serialize_model(model)
+        wrapped_future = wrap_future(training_future)
+        await wrapped_future
+        model_msg = wrapped_future.result()
         return model_msg
 
 
@@ -816,14 +458,23 @@ class DatabaseMonitor(CascadeAgent):
     def __init__(
         self,
         config: DatabaseMonitorConfig,
-        trainer: Handle[DummyTrainer],
-        dynamics_engine: Handle[DynamicsEngine]
+        trainer: Handle[Trainer],
+        dynamics_runners: list[Handle[DynamicsRunner]]
     ):
-        super().__init__(config)
+        self.db_url = config.db_url
+        self.config = config
+        super().__init__()
         self.trainer = trainer
-        self.dynamics_engine = dynamics_engine
+        self.dynamics_runners = dynamics_runners
         self.last_train_count = 0
         self.current_training_round = 0
+        self.model_version = 0
+
+        # pull out config vars that might change
+        self.chunk_size = self.config.chunk_size
+        self.retrain_len = self.config.retrain_len
+        self.retrain_fraction = self.config.retrain_fraction
+        self.retrain_min_frames = self.config.retrain_min_frames
 
     @loop
     async def monitor_completion(self, shutdown: asyncio.Event) -> None:
@@ -831,16 +482,16 @@ async def monitor_completion(self, shutdown: asyncio.Event) -> None:
         while not shutdown.is_set():
             # Check if all trajectories are complete
             trajectories = self._traj_db.list_trajectories_in_run(self.config.run_id)
-            
+
             if len(trajectories) == 0:
                 await asyncio.sleep(1)
                 continue
-            
+
             all_finished = all(
                 traj['status'] in (TrajectoryStatus.COMPLETED, TrajectoryStatus.FAILED)
                 for traj in trajectories
             )
-            
+
             if all_finished:
                 self.logger.info("All trajs done, setting shutdown")
                 shutdown.set()
@@ -848,75 +499,58 @@ async def monitor_completion(self, shutdown: asyncio.Event) -> None:
             else:
                 await asyncio.sleep(1)
 
-    def _log_and_check_resource_counts(self) -> None:
-        """Log resource creation counts and alert if counts are unexpectedly high"""
-        # Log resource creation counts
-        self.logger.info(
-            f"Resource counts: agents={CascadeAgent._resource_counts['agents_created']}, "
-            f"traj_db_instances={CascadeAgent._resource_counts['traj_db_instances_created']}"
-        )
-        
-        # Alert if resource counts are unexpectedly high
-        if CascadeAgent._resource_counts['traj_db_instances_created'] > CascadeAgent._resource_counts['agents_created'] * 2:
-            self.logger.warning(
-                f"More TrajectoryDB instances ({CascadeAgent._resource_counts['traj_db_instances_created']}) "
-                f"than expected for {CascadeAgent._resource_counts['agents_created']} agents. "
-                "Instances may be recreated unnecessarily."
-            )
-
     @loop
     async def periodic_retrain(self, shutdown: asyncio.Event) -> None:
         """Monitor for enough training frames and trigger retraining.
-        
+
         Retraining is triggered when either condition is met:
         - Absolute threshold: number of new training frames >= retrain_len
         - Fraction threshold: fraction of active trajectories with samples >= retrain_fraction
-        
+
         The absolute condition ensures retraining happens after accumulating a minimum
         number of frames, while the fraction condition ensures retraining occurs when
         a sufficient proportion of active trajectories have been sampled, even if the
         absolute count is low.
-        
+
         After retraining, frames are resubmitted for execution with the new model.
         """
         self.logger.info("periodic_retrain loop started")
         while not shutdown.is_set():
-            self._log_and_check_resource_counts()
-            
+            await asyncio.sleep(5)
             # Check if we have enough new training frames
             current_count = self._traj_db.count_training_frames(self.config.run_id)
             new_frames = current_count - self.last_train_count
-            
+
             # Check fraction-based condition
             total_active, active_with_labeling = self._traj_db.count_active_trajs_with_labeling(
                 run_id=self.config.run_id
             )
             sampled_fraction = active_with_labeling / total_active if total_active > 0 else 0.
-            
-            self.logger.info(
-                f"Retrain check: new={new_frames}, active={total_active}, labeled={active_with_labeling}, "
-                f"fraction={sampled_fraction:.2%}"
-            )
-            
+
             # Determine which condition triggered retraining
-            absolute_condition = new_frames >= self.config.retrain_len
-            fraction_condition = sampled_fraction >= self.config.retrain_fraction
+            absolute_condition = new_frames >= self.retrain_len
+            fraction_condition = sampled_fraction >= self.retrain_fraction
             should_retrain = absolute_condition or fraction_condition
-            
+
+            self.logger.info(
+                f"Retrain check: new training frames={new_frames}, active trajectories={total_active}, trajectories with labeled frames={active_with_labeling}, "
+                f"fraction={sampled_fraction:.2%}, should_retrain={should_retrain}"
+            )
+
             if should_retrain:
                 trigger_reason = []
                 if absolute_condition:
-                    trigger_reason.append(f"absolute threshold ({new_frames} >= {self.config.retrain_len})")
+                    trigger_reason.append(f"absolute threshold ({new_frames} >= {self.retrain_len})")
                 if fraction_condition:
-                    trigger_reason.append(f"fraction threshold ({sampled_fraction:.2%} >= {self.config.retrain_fraction:.2%})")
-                
+                    trigger_reason.append(f"fraction threshold ({sampled_fraction:.2%} >= {self.retrain_fraction:.2%})")
+
                 # Get the training round for frames that will be used in this retraining
                 # (frames created before this retraining will have the current max training_round)
                 training_round_for_retrain = self._traj_db.get_current_training_round(self.config.run_id)
-                
+
                 # Increment training round - new frames created after this will use the new round
                 self.current_training_round = training_round_for_retrain + 1
-                
+
                 self.logger.info(
                     f"Starting retraining (round {self.current_training_round}) triggered by: {', '.join(trigger_reason)}\n"
                     f"Training frame count: current={current_count}, last_train={self.last_train_count}, "
@@ -926,83 +560,18 @@ async def periodic_retrain(self, shutdown: asyncio.Event) -> None:
 
                 # Train model and update weights in dynamics engine
                 weights = await self.trainer.train_model(self.current_training_round)
-                
+
                 # Record FINISHED_TRAINING event after training completes
                 self._traj_db.record_training_event(
                     run_id=self.config.run_id,
                     event_type=ChunkEventType.FINISHED_TRAINING,
                     training_round=self.current_training_round
                 )
-                
-                await self.dynamics_engine.receive_weights(weights)
-                
-                # Get chunks where latest event is FINISHED_LABELING (these need resubmission)
-                chunks_to_resubmit = self._traj_db.get_trajs_with_latest_event(
-                    run_id=self.config.run_id,
-                    event_type=ChunkEventType.FINISHED_SAMPLING
-                )
-                self.logger.info(f"Found {len(chunks_to_resubmit)} trajs to resubmit (latest event FINISHED_LABELING)")
-
-                # These should all be FAILED chunks - resubmit the SAME chunk_id (will create new attempt)
-                for chunk_info in chunks_to_resubmit:
-                    traj_id = chunk_info['traj_id']
-                    chunk_id = chunk_info['chunk_id']
-                    attempt_index = chunk_info['attempt_index']
-                    
-                    # Get the starting frame for resubmission:
-                    # - If chunk_id > 0: use first frame of previous chunk (chunk_id - 1)
-                    # - If chunk_id == 0: use initial frame from trajectory
-                    if chunk_id > 0:
-                        # Get the latest attempt of the previous chunk
-                        # Since chunk N exists, chunk N-1 must have passed, so the latest attempt should be the passed one
-                        prev_chunk_attempt = self._traj_db.get_latest_chunk_attempt(
-                            run_id=self.config.run_id,
-                            traj_id=traj_id,
-                            chunk_id=chunk_id - 1
-                        )
-                        # Get last frame of previous chunk
-                        start_frame = self._traj_db.get_last_frame_from_chunk(
-                            run_id=self.config.run_id,
-                            traj_id=traj_id,
-                            chunk_id=chunk_id - 1,
-                            attempt_index=prev_chunk_attempt['attempt_index']
-                        )
-                    else:
-                        # chunk_id == 0: use initial trajectory frame
-                        start_frame = self._traj_db.get_initial_trajectory_frame(
-                            run_id=self.config.run_id,
-                            traj_id=traj_id
-                        )
-
-                    # Resubmit the SAME chunk_id (dynamics engine will create a new attempt)
-                    next_attempt_index = self._traj_db.get_next_attempt_index(
-                        run_id=self.config.run_id,
-                        traj_id=traj_id,
-                        chunk_id=chunk_id
-                    )
-                    
-                    if start_frame:
-                        retry_spec = AdvanceSpec(
-                            atoms=start_frame,
-                            run_id=self.config.run_id,
-                            traj_id=traj_id,
-                            chunk_id=chunk_id,  # Same chunk_id, not chunk_id + 1
-                            attempt_index=next_attempt_index,
-                            steps=self.config.chunk_size
-                        )
-                        await self.dynamics_engine.submit(retry_spec)
-                        self.logger.info(
-                            f"Resubmitted traj {traj_id}, chunk {chunk_id} for retry "
-                            f"(from attempt {attempt_index} to attempt {next_attempt_index})"
-                        )
-                    else:
-                        self.logger.warning(
-                            f"Traj {traj_id}: No starting frame found for chunk {chunk_id} attempt {next_attempt_index} "
-                            f"({'initial trajectory' if chunk_id == 0 else f'previous chunk {chunk_id - 1}'})"
-                        )
-                
-                self.last_train_count = current_count
-                self.logger.info(f"Retraining complete, setting frame count to {current_count}")
-            else:
-                self.logger.debug(f"Retraining not triggered, sleeping for 5 seconds")
-                await asyncio.sleep(5)
+
+                self.model_version += 1
+
+                for runner in self.dynamics_runners:
+                    try:
+                        await runner.receive_weights(weights, self.model_version)
+                    except AgentTerminatedError:
+                        pass
diff --git a/cascade/agents/config.py b/cascade/agents/config.py
index f6a4edd..c20ece7 100644
--- a/cascade/agents/config.py
+++ b/cascade/agents/config.py
@@ -4,10 +4,19 @@
 from typing import TYPE_CHECKING
 
 if TYPE_CHECKING:
+    from ase import Atoms
     from ase.optimize.optimize import Dynamics
     from cascade.model import Trajectory
     from cascade.learning.base import BaseLearnableForcefield
     import numpy as np
+    from concurrent.futures import Executor
+    from typing import Callable
+    from cascade.model import (
+        AdvanceSpec,
+        AuditResult,
+        ChunkSpec,
+        TrainingFrameSpec
+    )
 
 
 @dataclass
@@ -18,6 +27,7 @@ class CascadeAgentConfig:
     db_url: str
     """Database URL"""
 
+
 @dataclass
 class DatabaseConfig(CascadeAgentConfig):
     """Configuration for DummyDatabase agent"""
@@ -27,48 +37,74 @@ class DatabaseConfig(CascadeAgentConfig):
 
 
 @dataclass
-class DynamicsEngineConfig(CascadeAgentConfig):
+class DynamicsRunnerConfig(CascadeAgentConfig):
     """Configuration for DynamicsEngine agent"""
-    #init_specs: list  # list[AdvanceSpec], but avoid circular import
+    atoms: Atoms
+    run_id: str
+    db_url: str
+    traj_id: int
+    chunk_size: int
+    n_steps: int
+    run_dir: str
+    executor: Executor
+    advance_dynamics_task: Callable[[AdvanceSpec], None]
     learner: BaseLearnableForcefield
     weights: bytes
     dyn_cls: type[Dynamics]
     dyn_kws: dict[str, object] | None
     run_kws: dict[str, object] | None
     device: str = 'cpu'
+    model_version: int = 0
 
 
 @dataclass
 class AuditorConfig(CascadeAgentConfig):
     """Configuration for DummyAuditor agent"""
-    accept_rate: float
+    run_id: int
+    db_url: str
+    audit_task: Callable[[ChunkSpec], AuditResult]
+    audit_kwargs: dict
+    executor: Executor
     chunk_size: int
 
 
 @dataclass
 class SamplerConfig(CascadeAgentConfig):
     """Configuration for Sampler agent"""
+    run_id: str
+    db_url: str
     n_frames: int
-    rng: np.random.Generator | None = None
+    executor: Executor
+    sample_task: Callable[..., list[TrainingFrameSpec]]
 
 
 @dataclass
 class LabelerConfig(CascadeAgentConfig):
     """Configuration for DummyLabeler agent"""
+    run_id: str
+    db_url: str
+    executor: Executor
+    label_task: Callable[TrainingFrameSpec, TrainingFrameSpec]
 
 
 @dataclass
 class TrainerConfig(CascadeAgentConfig):
     """Configuration for DummyTrainer agent"""
+    run_id: str
+    db_url: str
+    training_task: Callable[..., bytes]  # todo: come up with a signature
+    training_args: list | tuple
+    training_kwargs: dict
     learner: BaseLearnableForcefield
+    executor: Executor
 
 
 @dataclass
 class DatabaseMonitorConfig(CascadeAgentConfig):
     """Configuration for DatabaseMonitor agent"""
+    run_id: int
+    db_url: str
     retrain_len: int
-    target_length: int
     chunk_size: int
     retrain_fraction: float = 0.5
     retrain_min_frames: int = 10
-
diff --git a/cascade/agents/task.py b/cascade/agents/task.py
index 5d62a88..c68f2cf 100644
--- a/cascade/agents/task.py
+++ b/cascade/agents/task.py
@@ -2,23 +2,24 @@
 
 from typing import TYPE_CHECKING
 if TYPE_CHECKING:
-    from cascade.model import ChunkSpec, AuditResult
+    from cascade.model import ChunkSpec, AuditResult, Chunk
     from cascade.model import AdvanceSpec, TrainingFrameSpec
     from cascade.learning.base import BaseLearnableForcefield
     from ase import Atoms
+    from pathlib import Path
+    import numpy as np
 from ase.optimize.optimize import Dynamics
 
+
 # can make this a classmethod on some audittask class
 # to get some shared informaiton and inheritance
 def random_audit(
-    chunk_atoms: list[Atoms],
-    chunk_spec: ChunkSpec,
-    attempt_index: int,
+    chunk: Chunk,
     accept_prob: float = 0.5,
     sleep_time: float = 0.,
 ) -> AuditResult:
     """Random audit of a chunk of a trajectory
-    
+
     Intended to be used as a stub for a real audit function.
     """
     from cascade.model import AuditResult, AuditStatus
@@ -32,17 +33,14 @@ def random_audit(
     passed = rng.random() < accept_prob
     score = rng.random() if passed else 0.0
     status = AuditStatus.PASSED if passed else AuditStatus.FAILED
-    return AuditResult(status=status, score=score, traj_id=chunk_spec.traj_id, chunk_id=chunk_spec.chunk_id, attempt_index=attempt_index)
+    return AuditResult(status=status, score=score, traj_id=chunk.traj_id, chunk_id=spec.chunk_id, attempt_index=attempt_index)
 
 
 def random_sample(
-    atoms_list: list[Atoms],
-    frame_ids: list[int],
-    chunk_spec: ChunkSpec,
-    model_version: int,
+    chunk: Chunk,
     n_frames: int,
     sleep_time: float = 0.,
-) -> list:
+) -> list[TrainingFrame]:
     """Random sample of frames from a chunk.
 
     Intended to be used as a stub for a real sampling function.
@@ -55,23 +53,16 @@ def random_sample(
     # Create a new random generator seeded with OS entropy to ensure
     # each worker process gets a unique random state
     rng = np.random.default_rng(seed=None)
-    n_sample = min(n_frames, len(atoms_list))
-    indices = rng.choice(len(atoms_list), size=n_sample, replace=False)
-    sampled_frames = [atoms_list[i] for i in indices]
-    sampled_frame_ids = [frame_ids[i] for i in indices]
-
+    n_sample = min(n_frames, len(chunk.atoms))
+    indices = rng.choice(len(chunk.atoms), size=n_sample, replace=False)
     result = []
-    for frame, trajectory_frame_id in zip(sampled_frames, sampled_frame_ids):
-        training_frame = TrainingFrame(atoms=frame, model_version=model_version)
-        spec = TrainingFrameSpec(
-            training_frame=training_frame,
-            trajectory_frame_id=trajectory_frame_id,
-            traj_id=chunk_spec.traj_id,
-            chunk_id=chunk_spec.chunk_id,
-            attempt_index=chunk_spec.attempt_index,
-            total_frames_in_chunk=n_sample,
+    for i in indices:
+        result.append(
+            TrainingFrame(
+                atoms=chunk.atoms[i],
+                model_version=chunk.model_version,
+            )
         )
-        result.append(spec)
     return result
 
 
@@ -80,50 +71,78 @@ def advance_dynamics(
     learner: BaseLearnableForcefield,
     weights: bytes,
     db_url: str,
-    device: str = 'cpu',
-    dyn_cls: type[Dynamics] = Dynamics,
-    dyn_kws: dict[str, object] = {},
-    run_kws: dict[str, object] = {},
-) -> None:
+    device: str,
+    run_dir: str,
+    dyn_cls: type[Dynamics],
+    dyn_kws: dict[str, object],
+    run_kws: dict[str, object],
+) -> list[Atoms]:
     """Advance dynamics of a chunk of a trajectory
-    
-    Intended to be used as a stub for a real advance dynamics function.
+
+    Arguments:
+        spec: contains atoms and metadata about trajectory
+        learner: used to make the calculator
+        weights: weights to add to the calculator
+        db_url: url to write frames to
+        device: for torch
+        dyn_cls: ASE dynamics class
+        dyn_kws: kws to the dynamics constructor
+        run_kws: kws to the dynamics run method
     """
     import numpy as np
     from cascade.utils import canonicalize
-    from cascade.agents.db_orm import TrajectoryDB
-    
-    # Create TrajectoryDB instance in the worker process
-    traj_db = TrajectoryDB(db_url)
-    
+    from pathlib import Path
+
+    import logging
+    import os
+
+    # todo: stop this from writing to the screen
+    logfile = str(Path(run_dir) / f'traj-{spec.traj_id}_chunk-{spec.chunk_id}_att-{spec.attempt_index}_md.log')
+    logger = logging.getLogger(logfile)
+    file_handler = logging.FileHandler(logfile)
+    formatter = logging.Formatter('%(asctime)s : %(levelname)s : %(name)s : %(message)s')
+    file_handler.setFormatter(formatter)
+    logger.addHandler(file_handler)
+
     atoms = spec.atoms
+    logger.info('Creating calculator')
     calc = learner.make_calculator(weights, device=device)
     atoms.calc = calc
 
+    logger.info('Creating dynamics class')
     dyn = dyn_cls(atoms, **dyn_kws)
-    
-    frame_index = 0  # Track frame index within this chunk
 
-    def write_to_db():
-        nonlocal frame_index
+    frames = []
+
+    def write_frame():
+        logger.info('getting results from calc')
         f = atoms.calc.results['forces']
         atoms.calc.results['forces'] = f.astype(np.float64)
         canonical_atoms = canonicalize(atoms)
-        
-        # Write frame to database
-        traj_db.write_frame(
-            run_id=spec.run_id,
-            traj_id=spec.traj_id,
-            chunk_id=spec.chunk_id,
-            attempt_index=spec.attempt_index,
-            frame_index=frame_index,
-            atoms=canonical_atoms
-        )
-        frame_index += 1
-    
-    dyn.attach(write_to_db)
 
+        logger.info('writing frame to db')
+        frames.append(canonical_atoms)
+
+    dyn.attach(write_frame)
+
+    logger.info('Starting dynamics')
     dyn.run(spec.steps, **run_kws)
-    
+    os.remove(logfile)
+
+    return frames
+
+
+def label_noop(spec: TrainingFrameSpec) -> TrainingFrameSpec:
+    """Returns forces from the training frame spec unmodified"""
     return spec
 
+
+# todo: this should be configurable, or at least not hard code magic knowledge
+def training_noop(learner: BaseLearnableForcefield) -> bytes:
+    """just return a model"""
+    from mace.calculators import mace_mp
+
+    calc = mace_mp('small', device='cpu', default_dtype="float32")
+    model = calc.models[0]
+    model_msg = learner.serialize_model(model)
+    return model_msg
\ No newline at end of file
diff --git a/cascade/model.py b/cascade/model.py
index 493e0e1..bdddfe2 100644
--- a/cascade/model.py
+++ b/cascade/model.py
@@ -6,6 +6,24 @@
 
 from ase import Atoms
 
+
+
+@dataclass
+class Chunk:
+    atoms: list[Atoms]
+    traj_id: int
+    chunk_id: int
+    attempt_ix: int
+    model_version: int
+
+
+@dataclass
+class AuditResult:
+    """The result of an audit"""
+    status: AuditStatus
+    """Whether the chunk passed audit"""
+    score: float
+
 @dataclass
 class ChunkSpec:
     traj_id: int
@@ -13,10 +31,14 @@ class ChunkSpec:
     attempt_index: int | None = None
     model_version: int | None = None
 
+
+# todo: rethink this frame/spec model. I am not sure this makes sense anymore
 @dataclass
 class TrainingFrame:
     atoms: Atoms
     model_version: int
+    labeled: bool = False
+
 
 @dataclass
 class TrainingFrameSpec:
@@ -38,19 +60,17 @@ class TrainingFrameSpec:
     total_frames_in_chunk: int
     """Total number of frames that will be labeled for this chunk"""
 
+
 @dataclass
-class AuditResult:
-    """The result of an audit"""
-    status: AuditStatus
-    """Whether the chunk passed audit"""
-    score: float
-    """The score assigned by the auditor"""
-    traj_id: int
-    """The trajectory ID"""
-    chunk_id: int
-    """The chunk ID""" 
-    attempt_index: int
-    """The attempt index"""
+class TrajectoryState:
+    atoms: Atoms
+    timestep: int
+    chunk: int
+    attempt: int
+
+
+
+
 
 class AuditStatus(Enum):
     """Whether a trajectory chunk is awaiting or has passed/failed an audit"""
@@ -58,12 +78,14 @@ class AuditStatus(Enum):
     FAILED = auto()
     PASSED = auto()
 
+
 class TrajectoryStatus(Enum):
     """Lifecycle state for a trajectory."""
     RUNNING = auto()
     COMPLETED = auto()
     FAILED = auto()
 
+
 class ChunkEventType(Enum):
     """Event types tracked for trajectory chunks"""
     STARTED_DYNAMICS = auto()
@@ -81,6 +103,7 @@ class ChunkEventType(Enum):
     STARTED_TRAINING = auto()
     FINISHED_TRAINING = auto()
 
+
 @dataclass
 class TrajectorySpec:
     """Enough information to initialize a trajectory"""
@@ -93,6 +116,7 @@ class TrajectorySpec:
     init_atoms: Atoms
     """The initial atoms for the trajectory"""
 
+
 @dataclass
 class AdvanceSpec:
     """Trajectory advancement specification.
diff --git a/tests/agents/test_dynamics_runner.py b/tests/agents/test_dynamics_runner.py
new file mode 100644
index 0000000..7a0ac2f
--- /dev/null
+++ b/tests/agents/test_dynamics_runner.py
@@ -0,0 +1,117 @@
+import pathlib
+import pytest
+from pytest import fixture
+from concurrent.futures import ThreadPoolExecutor, ProcessPoolExecutor
+import datetime
+
+from mace.calculators import mace_mp
+from ase import Atoms
+from ase.io import read
+from ase import units
+from ase.md.verlet import VelocityVerlet
+from academy.exchange import LocalExchangeFactory
+from academy.manager import Manager
+from academy.logging import init_logging
+
+from cascade.agents.agents import DynamicsRunner, CascadeAgent
+from cascade.learning.mace import MACEInterface
+from cascade.agents.db_orm import TrajectoryDB
+from cascade.model import AuditStatus, AdvanceSpec
+from cascade.agents.task import advance_dynamics
+
+
+class DeterministicAuditor(CascadeAgent):
+
+    def __init__(self,
+                 sequence=None
+                 ):
+        self.sequence = sequence
+
+    async def audit(self):
+        return self.sequence.pop(0)
+
+
+@fixture
+def atoms() -> Atoms:
+    return read('../../0_setup/final-geometries/packmol-CH4-in-H2O=32-seed=1-mace-medium.vasp')
+
+
+@pytest.mark.asyncio
+async def test_dynamics_runner(atoms: Atoms):
+    start_time = datetime.datetime.utcnow().strftime("%Y.%m.%d-%H:%M:%S")
+    run_id = f'test-{start_time}'
+    target_length = 10
+    chunk_size = 5
+    db_url = 'postgresql://ase:pw@localhost:5432/cascade'
+
+    run_dir = pathlib.Path("run") / run_id
+    run_dir.mkdir(parents=True)
+    learner = MACEInterface()
+    init_weights = learner.serialize_model(learner.get_model(mace_mp('small').models[0]))
+    logfile = run_dir / "runtime.log"
+    logger = init_logging(level='INFO', logfile=logfile)
+
+    # Initialize trajectories in the database
+    traj_db = TrajectoryDB(db_url)
+    traj_db.create_tables()
+    traj_db.initialize_trajectory(
+        run_id=run_id,
+        traj_id=0,
+        target_length=target_length,
+        init_atoms=atoms
+    )
+
+    # intial conditions, trajectories
+    initial_specs = []
+    initial_specs.append(
+        AdvanceSpec(
+            atoms=atoms,
+            run_id=run_id,
+            traj_id=0,
+            chunk_id=0,
+            attempt_index=0,
+            steps=chunk_size
+        )
+    )
+
+    # initialize manager, exchange
+    async with await Manager.from_exchange_factory(
+        factory=LocalExchangeFactory(),
+        executors=ThreadPoolExecutor(max_workers=10),
+    ) as manager:
+
+    dyn_reg = await manager.register_agent(DynamicsRunner)
+    aud_reg = await manager.register_agent(DeterministicAuditor)
+    dyn_handle = manager.get_handle(dyn_reg)
+    aud_handle = manager.get_handle(aud_reg)
+
+    await manager.launch(
+        DynamicsRunner,
+        kwargs=dict(
+            atoms=atoms,
+            run_id=run_id,
+            traj_id=0,
+            chunk_size=chunk_size,
+            n_steps=target_length,
+            auditor=aud_handle,
+            executor=ProcessPoolExecutor(1),
+            advance_dynamics_task=advance_dynamics,
+            learner=learner,
+            weights=init_weights,
+            dyn_cls=VelocityVerlet,
+            dyn_kws={'timestep': 1 * units.fs},
+            run_kws={},
+            device='cpu',
+            model_version=0
+        ),
+        registration=dyn_reg
+    )
+
+    await manager.launch(
+        DeterministicAuditor,
+        sequence=[AuditStatus.PASSED, AuditStatus.FAILED, AuditStatus.PASSED],
+    )
+
+    await manager.wait([dyn_handle]) # this should wait until it shuts itself down
+
+    # todo: use caplog to assert things happen as expected
\ No newline at end of file