Merge pull request #232 from Labelbox/develop

3.0.0rc2

Author: msokoloff1

CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+# Version 3.0.0-rc2 (2021-08-09)
+## Updates
+* Rename `data` property of TextData, ImageData, and VideoData types to `value`.
+* Decrease wait time between updates for `BulkImportRequest.wait_until_done()`
+* Organization is no longer used to create the LFO in `Project.setup()`
+
+
 # Version 3.0.0-rc1 (2021-08-05)
 ## Added
 * Model diagnostics notebooks

examples/annotation_types/basics.ipynb

@@ -33,7 +33,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install \"labelbox[data]==3.0.0rc1\""
+    "!pip install \"labelbox[data]\" --pre"
    ]
   },
   {
@@ -55,7 +55,7 @@
    "source": [
     "from labelbox.data.annotation_types import (\n",
     "    Label,\n",
-    "    RasterData,\n",
+    "    ImageData,\n",
     "    LabelList,\n",
     "    TextData,\n",
     "    VideoData,\n",
@@ -176,7 +176,7 @@
    "outputs": [],
    "source": [
     "labels = [Label(\n",
-    "    data = RasterData(url = \"http://my-img.jpg\"),\n",
+    "    data = ImageData(url = \"http://my-img.jpg\"),\n",
     "    annotations = [\n",
     "        ObjectAnnotation(\n",
     "            value = Point(x = 10, y = 10),\n",
@@ -215,7 +215,7 @@
     "* There are three classes to represent data\n",
     "    * TextData\n",
     "    * VideoData\n",
-    "    * RasterData\n",
+    "    * ImageData\n",
     "* The data objects can be constructed from various representations\n",
     "    - remote (url, uri)\n",
     "    - disk ( file path)\n",
@@ -229,7 +229,7 @@
    "id": "associate-cutting",
    "metadata": {},
    "source": [
-    "### RasterData"
+    "### ImageData"
    ]
   },
   {
@@ -240,13 +240,13 @@
    "outputs": [],
    "source": [
     "# Data can be instantiated with any of the following\n",
-    "image = RasterData(im_bytes = b'bytes')\n",
-    "image = RasterData(arr = np.zeros((10,10,3)).astype(np.uint8))\n",
-    "image = RasterData(file_path = '/tmp/img.jpg')\n",
+    "image = ImageData(im_bytes = b'bytes')\n",
+    "image = ImageData(arr = np.zeros((10,10,3)).astype(np.uint8))\n",
+    "image = ImageData(file_path = '/tmp/img.jpg')\n",
     "image_url = \"https://picsum.photos/id/1003/200/300\"\n",
-    "image = RasterData(url = image_url)\n",
+    "image = ImageData(url = image_url)\n",
     "# All of these can access the numpy representation the same way:\n",
-    "np_data = image.data\n",
+    "np_data = image.value\n",
     "\n",
     "\n",
     "im = Image.fromarray(np_data)\n",
@@ -273,7 +273,7 @@
     "text = TextData(text = \" some text content...\")\n",
     "text = TextData(url = \"https://filesamples.com/samples/document/txt/sample3.txt\")\n",
     "\n",
-    "print(text.data[:100])"
+    "print(text.value[:100])"
    ]
   },
   {
@@ -306,7 +306,7 @@
    "outputs": [],
    "source": [
     "\n",
-    "for idx, frame in video.data:\n",
+    "for idx, frame in video.value:\n",
     "    # Show every 50th frame\n",
     "    if idx % 50 == 0:\n",
     "        video_im = Image.fromarray(frame)\n",
@@ -342,7 +342,7 @@
    "outputs": [],
    "source": [
     "td = TextData(text = \"some text\", uid = \"ckrey6o07000008l50uk2gcr3\", external_id = \"my_text_data_row\")\n",
-    "rd = RasterData(url = image_url, uid = \"ckrey7te2000108l5hl8564dr\", external_id = \"my_raster_data_row\")\n",
+    "rd = ImageData(url = image_url, uid = \"ckrey7te2000108l5hl8564dr\", external_id = \"my_raster_data_row\")\n",
     "vd = VideoData(url = video_url, uid = \"ckrey7xef000208l57hwfgi3v\", external_id = \"my_video_data_row\")"
    ]
   },
@@ -539,7 +539,7 @@
    "metadata": {},
    "source": [
     "##### Mask\n",
-    "* The mask can be any RasterData object.\n"
+    "* The mask can be any ImageData object.\n"
    ]
   },
   {
@@ -551,12 +551,12 @@
    "source": [
     "\n",
     "mask_annotation = ObjectAnnotation(\n",
-    "        value = Mask(mask =  RasterData(arr = np.zeros((32,32,3), dtype = np.uint8)), color = (255,255,255)),\n",
+    "        value = Mask(mask =  ImageData(arr = np.zeros((32,32,3), dtype = np.uint8)), color = (255,255,255)),\n",
     "            name = \"mask class name\"\n",
     ")\n",
     "\n",
     "mask_annotation = ObjectAnnotation(\n",
-    "        value = Mask(mask =  RasterData(arr = np.zeros((32,32,3), dtype = np.uint8)), color = (255,255,255)),\n",
+    "        value = Mask(mask =  ImageData(arr = np.zeros((32,32,3), dtype = np.uint8)), color = (255,255,255)),\n",
     "            schema_id = \"ckrgcel71000008jtd9mn0szu\"\n",
     ")"
    ]
@@ -577,7 +577,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "raster_data = RasterData(arr = np.zeros((32,32,3), dtype = np.uint8))\n",
+    "raster_data = ImageData(arr = np.zeros((32,32,3), dtype = np.uint8))\n",
     "mask_annotation = ObjectAnnotation(\n",
     "        value = Mask(mask  = raster_data, color = [255,255,255]),\n",
     "            name = \"eyes\"\n",
@@ -604,7 +604,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "raster_data = RasterData(arr = np.zeros((32,32, 3), dtype = np.uint8))\n",
+    "raster_data = ImageData(arr = np.zeros((32,32, 3), dtype = np.uint8))\n",
     "mask_annotation = ObjectAnnotation(\n",
     "        value = Mask(mask  = raster_data, color = (128,255,255)),\n",
     "            name = \"eye\"\n",
@@ -734,7 +734,7 @@
    "outputs": [],
    "source": [
     "mask_annotation = Mask(\n",
-    "    mask = RasterData(arr = np_mask),\n",
+    "    mask = ImageData(arr = np_mask),\n",
     "    color = color \n",
     ")\n",
     "\n",
@@ -790,7 +790,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "mask_data = RasterData(arr = np_seg_mask)\n",
+    "mask_data = ImageData(arr = np_seg_mask)\n",
     "eye_mask = Mask(mask = mask_data, color = eye_color)\n",
     "nose_mask = Mask(mask = mask_data, color = nose_color)\n"
    ]
@@ -946,15 +946,15 @@
    "outputs": [],
    "source": [
     "label = Label(\n",
-    "    data = RasterData(url = image_url),\n",
+    "    data = ImageData(url = image_url),\n",
     "    annotations = [\n",
     "        ObjectAnnotation(\n",
     "            value = Polygon(points = [Point(x = x, y = y) for x,y in xy_poly]),\n",
     "            name = \"deer\"\n",
     "        ),\n",
     "        ObjectAnnotation(\n",
     "            name = \"deer_eyes\",\n",
-    "            value = Mask(mask = RasterData(arr = np_mask), color = color)\n",
+    "            value = Mask(mask = ImageData(arr = np_mask), color = color)\n",
     "        )\n",
     "    ]\n",
     ")"
@@ -994,7 +994,7 @@
    "outputs": [],
    "source": [
     "# Manually set just by adding a value:\n",
-    "image = RasterData(arr = np_data)\n",
+    "image = ImageData(arr = np_data)\n",
     "image.url = \"http://myurl\""
    ]
   },
@@ -1048,7 +1048,7 @@
    "source": [
     "\n",
     "label = Label(\n",
-    "    data = RasterData(arr = np_data),\n",
+    "    data = ImageData(arr = np_data),\n",
     "    annotations = [\n",
     "         ObjectAnnotation(\n",
     "            name = \"deer_nose\",\n",

examples/annotation_types/converters.ipynb

@@ -25,7 +25,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install \"labelbox[data]==3.0.0rc1\""
+    "!pip install \"labelbox[data]\" --pre"
    ]
   },
   {
@@ -361,7 +361,7 @@
     "\n",
     "for label in labels:\n",
     "    annotation_lookup = label.frame_annotations()\n",
-    "    for idx, frame in label.data.data:\n",
+    "    for idx, frame in label.data.value:\n",
     "        if idx % 30 != 1:\n",
     "            continue\n",
     "            \n",
@@ -412,7 +412,7 @@
     }
    ],
    "source": [
-    "im_data = label_list[0].data.data\n",
+    "im_data = label_list[0].data.value\n",
     "h,w = im_data.shape[:2]\n",
     "Image.fromarray(im_data)"
    ]

examples/annotation_types/label_containers.ipynb

@@ -20,7 +20,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install \"labelbox[data]==3.0.0rc1\""
+    "!pip install \"labelbox[data]\" --pre"
    ]
   },
   {
@@ -36,7 +36,7 @@
     "from labelbox import LabelingFrontend\n",
     "from labelbox.data.annotation_types import (\n",
     "    Label, \n",
-    "    RasterData, \n",
+    "    ImageData, \n",
     "    Mask, \n",
     "    Point, \n",
     "    Polygon, \n",
@@ -136,9 +136,9 @@
     "        Point(x = 99, y = 181).raster(im_h, im_w, thickness = 3),\n",
     "    ]\n",
     "    mask_arr = np.max([*eye_masks,nose_mask] , axis = 0)\n",
-    "    mask = RasterData(arr = mask_arr)\n",
+    "    mask = ImageData(arr = mask_arr)\n",
     "    return [Label(\n",
-    "        data = RasterData(im_bytes = requests.get(image_url).content),\n",
+    "        data = ImageData(im_bytes = requests.get(image_url).content),\n",
     "        annotations = [\n",
     "            ObjectAnnotation(value = get_polygon(),name = \"deer\"),\n",
     "            ObjectAnnotation(name = \"deer_eyes\", value = Mask(mask = mask, color = eye_color)),  \n",

examples/annotation_types/mal_using_annotation_types.ipynb

@@ -17,7 +17,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install \"labelbox[data]==3.0.0rc1\""
+    "!pip install \"labelbox[data]\" --pre"
    ]
   },
   {
@@ -66,7 +66,7 @@
     "\n",
     "from labelbox.data.annotation_types import (\n",
     "    LabelList,\n",
-    "    RasterData,\n",
+    "    ImageData,\n",
     "    Rectangle,\n",
     "    ObjectAnnotation,\n",
     "    ClassificationAnnotation,\n",
@@ -170,7 +170,7 @@
     "                start = Point(x = box[1], y = box[0]), end = Point(x = box[3], y = box[2])\n",
     "            )\n",
     "        elif name == 'car':\n",
-    "            value = Mask(mask = RasterData.from_2D_arr(arr = seg), color = (1,1,1))\n",
+    "            value = Mask(mask = ImageData.from_2D_arr(arr = seg), color = (1,1,1))\n",
     "        if value is not None:\n",
     "            annotations.append(\n",
     "                ObjectAnnotation(\n",
@@ -202,8 +202,8 @@
     "\n",
     "labellist = LabelList()\n",
     "for image_url in image_urls:\n",
-    "    image_data = RasterData(url = image_url)\n",
-    "    height, width = image_data.data.shape[:2]\n",
+    "    image_data = ImageData(url = image_url)\n",
+    "    height, width = image_data.value.shape[:2]\n",
     "    prediction = predict(np.array([image_data.im_bytes]), min_score=0.5, height=height, width = width)\n",
     "    annotations = get_annotations(prediction['boxes'], prediction['class_indices'], prediction['seg_masks'])\n",
     "    update_bag_classifications(annotations)\n",

examples/basics/data_rows.ipynb

@@ -180,7 +180,8 @@
    "id": "successful-patch",
    "metadata": {},
    "source": [
-    "### Create"
+    "### Create\n",
+    "* Create a single data row at a time"
    ]
   },
   {
@@ -201,7 +202,6 @@
     }
    ],
    "source": [
-    "#Add one at a time\n",
     "dataset = client.create_dataset(name=\"testing-dataset\")\n",
     "dataset.create_data_row(row_data=\"https://picsum.photos/200/300\")\n",
     "\n",
@@ -211,14 +211,21 @@
     "                        external_id=str(uuid.uuid4()))"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "helpful-lingerie",
+   "metadata": {},
+   "source": [
+    "* Bulk create data rows (This is much faster than creating individual data rows)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 10,
    "id": "round-perfume",
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Bulk create data_rows\n",
     "task1 = dataset.create_data_rows([{\n",
     "    DataRow.row_data: \"https://picsum.photos/200/300\"\n",
     "}, {\n",
@@ -382,7 +389,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },