From ee82af12f39bebe56e2d399815df77ce32700fba Mon Sep 17 00:00:00 2001 From: Aghin Shah Alin Date: Sun, 29 Dec 2019 10:39:47 +0530 Subject: [PATCH] add eval_mode to init of rainbow_agent --- dopamine/agents/rainbow/rainbow_agent.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/dopamine/agents/rainbow/rainbow_agent.py b/dopamine/agents/rainbow/rainbow_agent.py index bcc31856..b7436792 100644 --- a/dopamine/agents/rainbow/rainbow_agent.py +++ b/dopamine/agents/rainbow/rainbow_agent.py @@ -71,6 +71,7 @@ def __init__(self, epsilon_decay_period=250000, replay_scheme='prioritized', tf_device='/cpu:*', + eval_mode=False, use_staging=True, optimizer=tf.train.AdamOptimizer( learning_rate=0.00025, epsilon=0.0003125), @@ -110,6 +111,7 @@ def __init__(self, replay_scheme: str, 'prioritized' or 'uniform', the sampling scheme of the replay memory. tf_device: str, Tensorflow device on which the agent's graph is executed. + eval_mode: bool, True for evaluation and False for training. use_staging: bool, when True use a staging area to prefetch the next training batch, speeding training up by about 30%. optimizer: `tf.train.Optimizer`, for training the value function. @@ -144,6 +146,7 @@ def __init__(self, epsilon_eval=epsilon_eval, epsilon_decay_period=epsilon_decay_period, tf_device=tf_device, + eval_mode=eval_mode, use_staging=use_staging, optimizer=self.optimizer, summary_writer=summary_writer,