Train a Deep Q Network with TF-Agents

Copyright 2021 The TF-Agents Authors.

Introduction

This example shows how to train a DQN (Deep Q Networks) agent on the Cartpole environment using the TF-Agents library.

It will walk you through all the components in a Reinforcement Learning (RL) pipeline for training, evaluation and data collection.

To run this code live, click the 'Run in Google Colab' link above.

Setup

If you haven't installed the following dependencies, run:

sudo apt-get update
sudo apt-get install -y xvfb ffmpeg freeglut3-dev
pip install 'imageio==2.4.0'
pip install pyvirtualdisplay
pip install tf-agents[reverb]
pip install pyglet

from __future__ import absolute_import, division, print_function

import base64
import imageio
import IPython
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
import PIL.Image
import pyvirtualdisplay
import reverb

import tensorflow as tf

from tf_agents.agents.dqn import dqn_agent
from tf_agents.drivers import py_driver
from tf_agents.environments import suite_gym
from tf_agents.environments import tf_py_environment
from tf_agents.eval import metric_utils
from tf_agents.metrics import tf_metrics
from tf_agents.networks import sequential
from tf_agents.policies import py_tf_eager_policy
from tf_agents.policies import random_tf_policy
from tf_agents.replay_buffers import reverb_replay_buffer
from tf_agents.replay_buffers import reverb_utils
from tf_agents.trajectories import trajectory
from tf_agents.specs import tensor_spec
from tf_agents.utils import common

# Set up a virtual display for rendering OpenAI gym environments.
display = pyvirtualdisplay.Display(visible=0, size=(1400, 900)).start()

tf.version.VERSION

'2.8.0'

Hyperparameters

num_iterations = 20000 # @param {type:"integer"}

initial_collect_steps = 100  # @param {type:"integer"}
collect_steps_per_iteration =   1# @param {type:"integer"}
replay_buffer_max_length = 100000  # @param {type:"integer"}

batch_size = 64  # @param {type:"integer"}
learning_rate = 1e-3  # @param {type:"number"}
log_interval = 200  # @param {type:"integer"}

num_eval_episodes = 10  # @param {type:"integer"}
eval_interval = 1000  # @param {type:"integer"}

Environment

In Reinforcement Learning (RL), an environment represents the task or problem to be solved. Standard environments can be created in TF-Agents using tf_agents.environments suites. TF-Agents has suites for loading environments from sources such as the OpenAI Gym, Atari, and DM Control.

Load the CartPole environment from the OpenAI Gym suite.

env_name = 'CartPole-v0'
env = suite_gym.load(env_name)

You can render this environment to see how it looks. A free-swinging pole is attached to a cart. The goal is to move the cart right or left in order to keep the pole pointing up.

env.reset()
PIL.Image.fromarray(env.render())

The environment.step method takes an action in the environment and returns a TimeStep tuple containing the next observation of the environment and the reward for the action.

The time_step_spec() method returns the specification for the TimeStep tuple. Its observation attribute shows the shape of observations, the data types, and the ranges of allowed values. The reward attribute shows the same details for the reward.

print('Observation Spec:')
print(env.time_step_spec().observation)

Observation Spec:
BoundedArraySpec(shape=(4,), dtype=dtype('float32'), name='observation', minimum=[-4.8000002e+00 -3.4028235e+38 -4.1887903e-01 -3.4028235e+38], maximum=[4.8000002e+00 3.4028235e+38 4.1887903e-01 3.4028235e+38])

print('Reward Spec:')
print(env.time_step_spec().reward)

Reward Spec:
ArraySpec(shape=(), dtype=dtype('float32'), name='reward')

The action_spec() method returns the shape, data types, and allowed values of valid actions.

print('Action Spec:')
print(env.action_spec())

Action Spec:
BoundedArraySpec(shape=(), dtype=dtype('int64'), name='action', minimum=0, maximum=1)

In the Cartpole environment:

• observation is an array of 4 floats:
  • the position and velocity of the cart
  • the angular position and velocity of the pole
• reward is a scalar float value
• action is a scalar integer with only two possible values:
  • 0 — "move left"
  • 1 — "move right"

time_step = env.reset()
print('Time step:')
print(time_step)

action = np.array(1, dtype=np.int32)

next_time_step = env.step(action)
print('Next time step:')
print(next_time_step)

Time step:
TimeStep(
{'discount': array(1., dtype=float32),
 'observation': array([ 0.03982952,  0.03095708,  0.02307781, -0.0216785 ], dtype=float32),
 'reward': array(0., dtype=float32),
 'step_type': array(0, dtype=int32)})
Next time step:
TimeStep(
{'discount': array(1., dtype=float32),
 'observation': array([ 0.04044866,  0.2257406 ,  0.02264424, -0.3069917 ], dtype=float32),
 'reward': array(1., dtype=float32),
 'step_type': array(1, dtype=int32)})

Usually two environments are instantiated: one for training and one for evaluation.

train_py_env = suite_gym.load(env_name)
eval_py_env = suite_gym.load(env_name)

The Cartpole environment, like most environments, is written in pure Python. This is converted to TensorFlow using the TFPyEnvironment wrapper.

The original environment's API uses Numpy arrays. The TFPyEnvironment converts these to Tensors to make it compatible with Tensorflow agents and policies.

train_env = tf_py_environment.TFPyEnvironment(train_py_env)
eval_env = tf_py_environment.TFPyEnvironment(eval_py_env)

Agent

The algorithm used to solve an RL problem is represented by an Agent. TF-Agents provides standard implementations of a variety of Agents, including:

• DQN (used in this tutorial)
• REINFORCE
• DDPG
• TD3
• PPO
• SAC

The DQN agent can be used in any environment which has a discrete action space.

At the heart of a DQN Agent is a QNetwork, a neural network model that can learn to predict QValues (expected returns) for all actions, given an observation from the environment.

We will use tf_agents.networks. to create a QNetwork. The network will consist of a sequence of tf.keras.layers.Dense layers, where the final layer will have 1 output for each possible action.

fc_layer_params = (100, 50)
action_tensor_spec = tensor_spec.from_spec(env.action_spec())
num_actions = action_tensor_spec.maximum - action_tensor_spec.minimum + 1

# Define a helper function to create Dense layers configured with the right
# activation and kernel initializer.
def dense_layer(num_units):
  return tf.keras.layers.Dense(
      num_units,
      activation=tf.keras.activations.relu,
      kernel_initializer=tf.keras.initializers.VarianceScaling(
          scale=2.0, mode='fan_in', distribution='truncated_normal'))

# QNetwork consists of a sequence of Dense layers followed by a dense layer
# with `num_actions` units to generate one q_value per available action as
# its output.
dense_layers = [dense_layer(num_units) for num_units in fc_layer_params]
q_values_layer = tf.keras.layers.Dense(
    num_actions,
    activation=None,
    kernel_initializer=tf.keras.initializers.RandomUniform(
        minval=-0.03, maxval=0.03),
    bias_initializer=tf.keras.initializers.Constant(-0.2))
q_net = sequential.Sequential(dense_layers + [q_values_layer])

Now use tf_agents.agents.dqn.dqn_agent to instantiate a DqnAgent. In addition to the time_step_spec, action_spec and the QNetwork, the agent constructor also requires an optimizer (in this case, AdamOptimizer), a loss function, and an integer step counter.

optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)

train_step_counter = tf.Variable(0)

agent = dqn_agent.DqnAgent(
    train_env.time_step_spec(),
    train_env.action_spec(),
    q_network=q_net,
    optimizer=optimizer,
    td_errors_loss_fn=common.element_wise_squared_loss,
    train_step_counter=train_step_counter)

agent.initialize()

Policies

A policy defines the way an agent acts in an environment. Typically, the goal of reinforcement learning is to train the underlying model until the policy produces the desired outcome.

In this tutorial:

• The desired outcome is keeping the pole balanced upright over the cart.
• The policy returns an action (left or right) for each time_step observation.

Agents contain two policies:

• agent.policy — The main policy that is used for evaluation and deployment.
• agent.collect_policy — A second policy that is used for data collection.

eval_policy = agent.policy
collect_policy = agent.collect_policy

Policies can be created independently of agents. For example, use tf_agents.policies.random_tf_policy to create a policy which will randomly select an action for each time_step.

random_policy = random_tf_policy.RandomTFPolicy(train_env.time_step_spec(),
                                                train_env.action_spec())

To get an action from a policy, call the policy.action(time_step) method. The time_step contains the observation from the environment. This method returns a PolicyStep, which is a named tuple with three components:

• action — the action to be taken (in this case, 0 or 1)
• state — used for stateful (that is, RNN-based) policies
• info — auxiliary data, such as log probabilities of actions

example_environment = tf_py_environment.TFPyEnvironment(
    suite_gym.load('CartPole-v0'))

time_step = example_environment.reset()

random_policy.action(time_step)

PolicyStep(action=<tf.Tensor: shape=(1,), dtype=int64, numpy=array([0])>, state=(), info=())

Metrics and Evaluation

The most common metric used to evaluate a policy is the average return. The return is the sum of rewards obtained while running a policy in an environment for an episode. Several episodes are run, creating an average return.

The following function computes the average return of a policy, given the policy, environment, and a number of episodes.

def compute_avg_return(environment, policy, num_episodes=10):

  total_return = 0.0
  for _ in range(num_episodes):

    time_step = environment.reset()
    episode_return = 0.0

    while not time_step.is_last():
      action_step = policy.action(time_step)
      time_step = environment.step(action_step.action)
      episode_return += time_step.reward
    total_return += episode_return

  avg_return = total_return / num_episodes
  return avg_return.numpy()[0]


# See also the metrics module for standard implementations of different metrics.
# https://github.com/tensorflow/agents/tree/master/tf_agents/metrics

Running this computation on the random_policy shows a baseline performance in the environment.

compute_avg_return(eval_env, random_policy, num_eval_episodes)

19.4

Replay Buffer

In order to keep track of the data collected from the environment, we will use Reverb, an efficient, extensible, and easy-to-use replay system by Deepmind. It stores experience data when we collect trajectories and is consumed during training.

This replay buffer is constructed using specs describing the tensors that are to be stored, which can be obtained from the agent using agent.collect_data_spec.

table_name = 'uniform_table'
replay_buffer_signature = tensor_spec.from_spec(
      agent.collect_data_spec)
replay_buffer_signature = tensor_spec.add_outer_dim(
    replay_buffer_signature)

table = reverb.Table(
    table_name,
    max_size=replay_buffer_max_length,
    sampler=reverb.selectors.Uniform(),
    remover=reverb.selectors.Fifo(),
    rate_limiter=reverb.rate_limiters.MinSize(1),
    signature=replay_buffer_signature)

reverb_server = reverb.Server([table])

replay_buffer = reverb_replay_buffer.ReverbReplayBuffer(
    agent.collect_data_spec,
    table_name=table_name,
    sequence_length=2,
    local_server=reverb_server)

rb_observer = reverb_utils.ReverbAddTrajectoryObserver(
  replay_buffer.py_client,
  table_name,
  sequence_length=2)

[reverb/cc/platform/tfrecord_checkpointer.cc:150]  Initializing TFRecordCheckpointer in /tmp/tmpdoioyfy9.
[reverb/cc/platform/tfrecord_checkpointer.cc:386] Loading latest checkpoint from /tmp/tmpdoioyfy9
[reverb/cc/platform/default/server.cc:71] Started replay server on port 24185

For most agents, collect_data_spec is a named tuple called Trajectory, containing the specs for observations, actions, rewards, and other items.

agent.collect_data_spec

Trajectory(
{'action': BoundedTensorSpec(shape=(), dtype=tf.int64, name='action', minimum=array(0), maximum=array(1)),
 'discount': BoundedTensorSpec(shape=(), dtype=tf.float32, name='discount', minimum=array(0., dtype=float32), maximum=array(1., dtype=float32)),
 'next_step_type': TensorSpec(shape=(), dtype=tf.int32, name='step_type'),
 'observation': BoundedTensorSpec(shape=(4,), dtype=tf.float32, name='observation', minimum=array([-4.8000002e+00, -3.4028235e+38, -4.1887903e-01, -3.4028235e+38],
      dtype=float32), maximum=array([4.8000002e+00, 3.4028235e+38, 4.1887903e-01, 3.4028235e+38],
      dtype=float32)),
 'policy_info': (),
 'reward': TensorSpec(shape=(), dtype=tf.float32, name='reward'),
 'step_type': TensorSpec(shape=(), dtype=tf.int32, name='step_type')})

agent.collect_data_spec._fields

('step_type',
 'observation',
 'action',
 'policy_info',
 'next_step_type',
 'reward',
 'discount')

Data Collection

Now execute the random policy in the environment for a few steps, recording the data in the replay buffer.

Here we are using 'PyDriver' to run the experience collecting loop. You can learn more about TF Agents driver in our drivers tutorial.

py_driver.PyDriver(
    env,
    py_tf_eager_policy.PyTFEagerPolicy(
      random_policy, use_tf_function=True),
    [rb_observer],
    max_steps=initial_collect_steps).run(train_py_env.reset())

(TimeStep(
 {'discount': array(1., dtype=float32),
  'observation': array([-0.01682967, -0.5797006 ,  0.11501301,  0.994641  ], dtype=float32),
  'reward': array(1., dtype=float32),
  'step_type': array(1, dtype=int32)}),
 ())

The replay buffer is now a collection of Trajectories.

# For the curious:
# Uncomment to peel one of these off and inspect it.
# iter(replay_buffer.as_dataset()).next()

The agent needs access to the replay buffer. This is provided by creating an iterable tf.data.Dataset pipeline which will feed data to the agent.

Each row of the replay buffer only stores a single observation step. But since the DQN Agent needs both the current and next observation to compute the loss, the dataset pipeline will sample two adjacent rows for each item in the batch (num_steps=2).

This dataset is also optimized by running parallel calls and prefetching data.

# Dataset generates trajectories with shape [Bx2x...]
dataset = replay_buffer.as_dataset(
    num_parallel_calls=3,
    sample_batch_size=batch_size,
    num_steps=2).prefetch(3)

dataset

<PrefetchDataset element_spec=(Trajectory(
{'action': TensorSpec(shape=(64, 2), dtype=tf.int64, name=None),
 'discount': TensorSpec(shape=(64, 2), dtype=tf.float32, name=None),
 'next_step_type': TensorSpec(shape=(64, 2), dtype=tf.int32, name=None),
 'observation': TensorSpec(shape=(64, 2, 4), dtype=tf.float32, name=None),
 'policy_info': (),
 'reward': TensorSpec(shape=(64, 2), dtype=tf.float32, name=None),
 'step_type': TensorSpec(shape=(64, 2), dtype=tf.int32, name=None)}), SampleInfo(key=TensorSpec(shape=(64, 2), dtype=tf.uint64, name=None), probability=TensorSpec(shape=(64, 2), dtype=tf.float64, name=None), table_size=TensorSpec(shape=(64, 2), dtype=tf.int64, name=None), priority=TensorSpec(shape=(64, 2), dtype=tf.float64, name=None)))>

iterator = iter(dataset)
print(iterator)

<tensorflow.python.data.ops.iterator_ops.OwnedIterator object at 0x7fac502b7750>

# For the curious:
# Uncomment to see what the dataset iterator is feeding to the agent.
# Compare this representation of replay data 
# to the collection of individual trajectories shown earlier.

# iterator.next()

Training the agent

Two things must happen during the training loop:

• collect data from the environment
• use that data to train the agent's neural network(s)

This example also periodicially evaluates the policy and prints the current score.

The following will take ~5 minutes to run.

try:
  %%time
except:
  pass

# (Optional) Optimize by wrapping some of the code in a graph using TF function.
agent.train = common.function(agent.train)

# Reset the train step.
agent.train_step_counter.assign(0)

# Evaluate the agent's policy once before training.
avg_return = compute_avg_return(eval_env, agent.policy, num_eval_episodes)
returns = [avg_return]

# Reset the environment.
time_step = train_py_env.reset()

# Create a driver to collect experience.
collect_driver = py_driver.PyDriver(
    env,
    py_tf_eager_policy.PyTFEagerPolicy(
      agent.collect_policy, use_tf_function=True),
    [rb_observer],
    max_steps=collect_steps_per_iteration)

for _ in range(num_iterations):

  # Collect a few steps and save to the replay buffer.
  time_step, _ = collect_driver.run(time_step)

  # Sample a batch of data from the buffer and update the agent's network.
  experience, unused_info = next(iterator)
  train_loss = agent.train(experience).loss

  step = agent.train_step_counter.numpy()

  if step % log_interval == 0:
    print('step = {0}: loss = {1}'.format(step, train_loss))

  if step % eval_interval == 0:
    avg_return = compute_avg_return(eval_env, agent.policy, num_eval_episodes)
    print('step = {0}: Average Return = {1}'.format(step, avg_return))
    returns.append(avg_return)

WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py:1082: calling foldr_v2 (from tensorflow.python.ops.functional_ops) with back_prop=False is deprecated and will be removed in a future version.
Instructions for updating:
back_prop=False is deprecated. Consider using tf.stop_gradient instead.
Instead of:
results = tf.foldr(fn, elems, back_prop=False)
Use:
results = tf.nest.map_structure(tf.stop_gradient, tf.foldr(fn, elems))
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
[reverb/cc/client.cc:165] Sampler and server are owned by the same process (9862) so Table uniform_table is accessed directly without gRPC.
step = 200: loss = 16.015960693359375
step = 400: loss = 47.185611724853516
step = 600: loss = 736.9923095703125
step = 800: loss = 175.4927215576172
step = 1000: loss = 65.5635757446289
step = 1000: Average Return = 171.1999969482422
step = 1200: loss = 1594.881103515625
step = 1400: loss = 32.85587692260742
step = 1600: loss = 88.44110107421875
step = 1800: loss = 2868.479248046875
step = 2000: loss = 1537.3463134765625
step = 2000: Average Return = 129.6999969482422
step = 2200: loss = 282.4945983886719
step = 2400: loss = 918.4564208984375
step = 2600: loss = 118.39675903320312
step = 2800: loss = 86.51387023925781
step = 3000: loss = 43.96769714355469
step = 3000: Average Return = 151.39999389648438
step = 3200: loss = 37.7552490234375
step = 3400: loss = 55.306640625
step = 3600: loss = 145.16598510742188
step = 3800: loss = 30.914730072021484
step = 4000: loss = 115.53938293457031
step = 4000: Average Return = 180.10000610351562
step = 4200: loss = 55.767601013183594
step = 4400: loss = 59.72187805175781
step = 4600: loss = 54.24274826049805
step = 4800: loss = 218.40969848632812
step = 5000: loss = 310.97296142578125
step = 5000: Average Return = 200.0
step = 5200: loss = 74.71685791015625
step = 5400: loss = 505.34075927734375
step = 5600: loss = 349.7851257324219
step = 5800: loss = 556.5531005859375
step = 6000: loss = 393.497802734375
step = 6000: Average Return = 200.0
step = 6200: loss = 632.537109375
step = 6400: loss = 233.87774658203125
step = 6600: loss = 149.68161010742188
step = 6800: loss = 771.809326171875
step = 7000: loss = 136.535888671875
step = 7000: Average Return = 200.0
step = 7200: loss = 214.4747314453125
step = 7400: loss = 2745.0703125
step = 7600: loss = 4504.373046875
step = 7800: loss = 295.11859130859375
step = 8000: loss = 1483.1090087890625
step = 8000: Average Return = 200.0
step = 8200: loss = 13224.90625
step = 8400: loss = 731.6187744140625
step = 8600: loss = 2498.115234375
step = 8800: loss = 2752.463134765625
step = 9000: loss = 571.6118774414062
step = 9000: Average Return = 194.3000030517578
step = 9200: loss = 433.17413330078125
step = 9400: loss = 494.7076416015625
step = 9600: loss = 409.4160461425781
step = 9800: loss = 444.74029541015625
step = 10000: loss = 417.860595703125
step = 10000: Average Return = 200.0
step = 10200: loss = 241.61526489257812
step = 10400: loss = 323.7486267089844
step = 10600: loss = 1961.7996826171875
step = 10800: loss = 467.81781005859375
step = 11000: loss = 75.74491882324219
step = 11000: Average Return = 200.0
step = 11200: loss = 652.3634643554688
step = 11400: loss = 211.48666381835938
step = 11600: loss = 170.49517822265625
step = 11800: loss = 119.90447998046875
step = 12000: loss = 991.9341430664062
step = 12000: Average Return = 200.0
step = 12200: loss = 101.59536743164062
step = 12400: loss = 148.53955078125
step = 12600: loss = 191.56800842285156
step = 12800: loss = 115.52975463867188
step = 13000: loss = 70.33795166015625
step = 13000: Average Return = 200.0
step = 13200: loss = 70.56562805175781
step = 13400: loss = 159.9202880859375
step = 13600: loss = 1406.8662109375
step = 13800: loss = 85.49430847167969
step = 14000: loss = 84.74932861328125
step = 14000: Average Return = 200.0
step = 14200: loss = 317.33819580078125
step = 14400: loss = 707.352294921875
step = 14600: loss = 53.67301940917969
step = 14800: loss = 100.62562561035156
step = 15000: loss = 764.060546875
step = 15000: Average Return = 200.0
step = 15200: loss = 90.79063415527344
step = 15400: loss = 122.30677795410156
step = 15600: loss = 69.46464538574219
step = 15800: loss = 554.35498046875
step = 16000: loss = 96.95709228515625
step = 16000: Average Return = 200.0
step = 16200: loss = 1974.1376953125
step = 16400: loss = 203.4429931640625
step = 16600: loss = 254.4991912841797
step = 16800: loss = 237.65127563476562
step = 17000: loss = 1236.594482421875
step = 17000: Average Return = 200.0
step = 17200: loss = 2782.07958984375
step = 17400: loss = 469.6650085449219
step = 17600: loss = 7143.4326171875
step = 17800: loss = 15037.6240234375
step = 18000: loss = 4602.244140625
step = 18000: Average Return = 200.0
step = 18200: loss = 264.796142578125
step = 18400: loss = 236.1230010986328
step = 18600: loss = 2736.951416015625
step = 18800: loss = 239.957275390625
step = 19000: loss = 534.1686401367188
step = 19000: Average Return = 200.0
step = 19200: loss = 391.1002197265625
step = 19400: loss = 352.6400146484375
step = 19600: loss = 232.7955780029297
step = 19800: loss = 140.59347534179688
step = 20000: loss = 22495.046875
step = 20000: Average Return = 200.0

Visualization

Plots

Use matplotlib.pyplot to chart how the policy improved during training.

One iteration of Cartpole-v0 consists of 200 time steps. The environment gives a reward of +1 for each step the pole stays up, so the maximum return for one episode is 200. The charts shows the return increasing towards that maximum each time it is evaluated during training. (It may be a little unstable and not increase monotonically each time.)

iterations = range(0, num_iterations + 1, eval_interval)
plt.plot(iterations, returns)
plt.ylabel('Average Return')
plt.xlabel('Iterations')
plt.ylim(top=250)

(21.814999198913576, 250.0)

Videos

Charts are nice. But more exciting is seeing an agent actually performing a task in an environment.

First, create a function to embed videos in the notebook.

def embed_mp4(filename):
  """Embeds an mp4 file in the notebook."""
  video = open(filename,'rb').read()
  b64 = base64.b64encode(video)
  tag = '''
  <video width="640" height="480" controls>
    <source src="data:video/mp4;base64,{0}" type="video/mp4">
  Your browser does not support the video tag.
  </video>'''.format(b64.decode())

  return IPython.display.HTML(tag)

Now iterate through a few episodes of the Cartpole game with the agent. The underlying Python environment (the one "inside" the TensorFlow environment wrapper) provides a render() method, which outputs an image of the environment state. These can be collected into a video.

def create_policy_eval_video(policy, filename, num_episodes=5, fps=30):
  filename = filename + ".mp4"
  with imageio.get_writer(filename, fps=fps) as video:
    for _ in range(num_episodes):
      time_step = eval_env.reset()
      video.append_data(eval_py_env.render())
      while not time_step.is_last():
        action_step = policy.action(time_step)
        time_step = eval_env.step(action_step.action)
        video.append_data(eval_py_env.render())
  return embed_mp4(filename)

create_policy_eval_video(agent.policy, "trained-agent")

WARNING:root:IMAGEIO FFMPEG_WRITER WARNING: input image is not divisible by macro_block_size=16, resizing from (400, 600) to (400, 608) to ensure video compatibility with most codecs and players. To prevent resizing, make your input image divisible by the macro_block_size or set the macro_block_size to None (risking incompatibility). You may also see a FFMPEG warning concerning speedloss due to data not being aligned.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/imageio/plugins/ffmpeg.py:727: DeprecationWarning: tostring() is deprecated. Use tobytes() instead.
  self._proc.stdin.write(im.tostring())
[swscaler @ 0x561c2ab733c0] Warning: data is not aligned! This can lead to a speed loss

For fun, compare the trained agent (above) to an agent moving randomly. (It does not do as well.)

create_policy_eval_video(random_policy, "random-agent")

WARNING:root:IMAGEIO FFMPEG_WRITER WARNING: input image is not divisible by macro_block_size=16, resizing from (400, 600) to (400, 608) to ensure video compatibility with most codecs and players. To prevent resizing, make your input image divisible by the macro_block_size or set the macro_block_size to None (risking incompatibility). You may also see a FFMPEG warning concerning speedloss due to data not being aligned.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/imageio/plugins/ffmpeg.py:727: DeprecationWarning: tostring() is deprecated. Use tobytes() instead.
  self._proc.stdin.write(im.tostring())
[swscaler @ 0x5558c4e7f3c0] Warning: data is not aligned! This can lead to a speed loss