Train a Deep Q Network with TF-Agents

View on TensorFlow.org Run in Google Colab View source on GitHub Download notebook

Introduction

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

Cartpole environment

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 install -y xvfb ffmpeg
pip install -q 'imageio==2.4.0'
pip install -q pyvirtualdisplay
pip install -q tf-agents
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 tensorflow as tf

from tf_agents.agents.dqn import dqn_agent
from tf_agents.drivers import dynamic_step_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 q_network
from tf_agents.policies import random_tf_policy
from tf_agents.replay_buffers import tf_uniform_replay_buffer
from tf_agents.trajectories import trajectory
from tf_agents.utils import common
tf.compat.v1.enable_v2_behavior()

# Set up a virtual display for rendering OpenAI gym environments.
display = pyvirtualdisplay.Display(visible=0, size=(1400, 900)).start()
tf.version.VERSION
'2.4.0'

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())

png

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(step_type=array(0, dtype=int32), reward=array(0., dtype=float32), discount=array(1., dtype=float32), observation=array([-0.02220688,  0.02916698, -0.03515396, -0.03343702], dtype=float32))
Next time step:
TimeStep(step_type=array(1, dtype=int32), reward=array(1., dtype=float32), discount=array(1., dtype=float32), observation=array([-0.02162354,  0.22477496, -0.03582269, -0.33700085], dtype=float32))

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:

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.

Use tf_agents.networks.q_network to create a QNetwork, passing in the observation_spec, action_spec, and a tuple describing the number and size of the model's hidden layers.

fc_layer_params = (100,)

q_net = q_network.QNetwork(
    train_env.observation_spec(),
    train_env.action_spec(),
    fc_layer_params=fc_layer_params)

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.compat.v1.train.AdamOptimizer(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([1])>, 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)
19.3

Replay Buffer

The replay buffer keeps track of data collected from the environment. This tutorial uses tf_agents.replay_buffers.tf_uniform_replay_buffer.TFUniformReplayBuffer, as it is the most common.

The constructor requires the specs for the data it will be collecting. This is available from the agent using the collect_data_spec method. The batch size and maximum buffer length are also required.

replay_buffer = tf_uniform_replay_buffer.TFUniformReplayBuffer(
    data_spec=agent.collect_data_spec,
    batch_size=train_env.batch_size,
    max_length=replay_buffer_max_length)

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(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)), action=BoundedTensorSpec(shape=(), dtype=tf.int64, name='action', minimum=array(0), maximum=array(1)), policy_info=(), next_step_type=TensorSpec(shape=(), dtype=tf.int32, name='step_type'), reward=TensorSpec(shape=(), dtype=tf.float32, name='reward'), discount=BoundedTensorSpec(shape=(), dtype=tf.float32, name='discount', minimum=array(0., dtype=float32), maximum=array(1., dtype=float32)))
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.

def collect_step(environment, policy, buffer):
  time_step = environment.current_time_step()
  action_step = policy.action(time_step)
  next_time_step = environment.step(action_step.action)
  traj = trajectory.from_transition(time_step, action_step, next_time_step)

  # Add trajectory to the replay buffer
  buffer.add_batch(traj)

def collect_data(env, policy, buffer, steps):
  for _ in range(steps):
    collect_step(env, policy, buffer)

collect_data(train_env, random_policy, replay_buffer, initial_collect_steps)

# This loop is so common in RL, that we provide standard implementations. 
# For more details see the drivers module.
# https://www.tensorflow.org/agents/api_docs/python/tf_agents/drivers

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
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow/python/autograph/operators/control_flow.py:1218: ReplayBuffer.get_next (from tf_agents.replay_buffers.replay_buffer) is deprecated and will be removed in a future version.
Instructions for updating:
Use `as_dataset(..., single_deterministic_pass=False) instead.

<PrefetchDataset shapes: (Trajectory(step_type=(64, 2), observation=(64, 2, 4), action=(64, 2), policy_info=(), next_step_type=(64, 2), reward=(64, 2), discount=(64, 2)), BufferInfo(ids=(64, 2), probabilities=(64,))), types: (Trajectory(step_type=tf.int32, observation=tf.float32, action=tf.int64, policy_info=(), next_step_type=tf.int32, reward=tf.float32, discount=tf.float32), BufferInfo(ids=tf.int64, probabilities=tf.float32))>
iterator = iter(dataset)

print(iterator)
<tensorflow.python.data.ops.iterator_ops.OwnedIterator object at 0x7f424ef1e080>

# 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]

for _ in range(num_iterations):

  # Collect a few steps using collect_policy and save to the replay buffer.
  collect_data(train_env, agent.collect_policy, replay_buffer, collect_steps_per_iteration)

  # 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/python3.6/site-packages/tensorflow/python/util/dispatch.py:201: 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))
step = 200: loss = 12.070486068725586
step = 400: loss = 8.28309440612793
step = 600: loss = 31.025835037231445
step = 800: loss = 13.815164566040039
step = 1000: loss = 8.706375122070312
step = 1000: Average Return = 146.10000610351562
step = 1200: loss = 5.937136650085449
step = 1400: loss = 8.289731979370117
step = 1600: loss = 33.76885223388672
step = 1800: loss = 47.953025817871094
step = 2000: loss = 52.09583282470703
step = 2000: Average Return = 27.799999237060547
step = 2200: loss = 5.141240119934082
step = 2400: loss = 21.2930908203125
step = 2600: loss = 25.088130950927734
step = 2800: loss = 56.888206481933594
step = 3000: loss = 76.79216003417969
step = 3000: Average Return = 21.700000762939453
step = 3200: loss = 26.36425018310547
step = 3400: loss = 4.469114780426025
step = 3600: loss = 34.6283073425293
step = 3800: loss = 14.598098754882812
step = 4000: loss = 17.743749618530273
step = 4000: Average Return = 46.0
step = 4200: loss = 26.917938232421875
step = 4400: loss = 56.554386138916016
step = 4600: loss = 19.315950393676758
step = 4800: loss = 16.675647735595703
step = 5000: loss = 136.85499572753906
step = 5000: Average Return = 126.4000015258789
step = 5200: loss = 35.19789123535156
step = 5400: loss = 88.04693603515625
step = 5600: loss = 70.27599334716797
step = 5800: loss = 72.77832794189453
step = 6000: loss = 57.98759078979492
step = 6000: Average Return = 127.4000015258789
step = 6200: loss = 191.23748779296875
step = 6400: loss = 6.994782447814941
step = 6600: loss = 90.5509033203125
step = 6800: loss = 38.94111251831055
step = 7000: loss = 8.685359001159668
step = 7000: Average Return = 130.5
step = 7200: loss = 128.6968536376953
step = 7400: loss = 82.48645782470703
step = 7600: loss = 44.56972122192383
step = 7800: loss = 55.02344512939453
step = 8000: loss = 141.50399780273438
step = 8000: Average Return = 159.3000030517578
step = 8200: loss = 6.795340061187744
step = 8400: loss = 13.114398956298828
step = 8600: loss = 91.94058990478516
step = 8800: loss = 82.01240539550781
step = 9000: loss = 44.955650329589844
step = 9000: Average Return = 178.8000030517578
step = 9200: loss = 68.20870208740234
step = 9400: loss = 154.3193359375
step = 9600: loss = 137.314453125
step = 9800: loss = 74.73216247558594
step = 10000: loss = 91.0711441040039
step = 10000: Average Return = 184.6999969482422
step = 10200: loss = 85.52545166015625
step = 10400: loss = 8.654191970825195
step = 10600: loss = 22.928178787231445
step = 10800: loss = 20.545730590820312
step = 11000: loss = 271.4393005371094
step = 11000: Average Return = 174.3000030517578
step = 11200: loss = 9.628021240234375
step = 11400: loss = 137.62472534179688
step = 11600: loss = 5.842207908630371
step = 11800: loss = 174.1510772705078
step = 12000: loss = 7.528541564941406
step = 12000: Average Return = 182.1999969482422
step = 12200: loss = 8.867502212524414
step = 12400: loss = 81.00005340576172
step = 12600: loss = 12.920578002929688
step = 12800: loss = 142.42262268066406
step = 13000: loss = 143.52105712890625
step = 13000: Average Return = 155.5
step = 13200: loss = 297.1153259277344
step = 13400: loss = 390.66412353515625
step = 13600: loss = 142.9129638671875
step = 13800: loss = 63.83035659790039
step = 14000: loss = 21.4559268951416
step = 14000: Average Return = 162.8000030517578
step = 14200: loss = 129.82550048828125
step = 14400: loss = 16.236011505126953
step = 14600: loss = 9.619779586791992
step = 14800: loss = 184.36654663085938
step = 15000: loss = 10.463872909545898
step = 15000: Average Return = 183.0
step = 15200: loss = 33.84937286376953
step = 15400: loss = 30.32630157470703
step = 15600: loss = 453.1896667480469
step = 15800: loss = 215.66067504882812
step = 16000: loss = 524.3294677734375
step = 16000: Average Return = 188.5
step = 16200: loss = 25.474245071411133
step = 16400: loss = 493.87481689453125
step = 16600: loss = 44.63227462768555
step = 16800: loss = 22.07838249206543
step = 17000: loss = 1202.003662109375
step = 17000: Average Return = 196.8000030517578
step = 17200: loss = 253.89215087890625
step = 17400: loss = 256.4725646972656
step = 17600: loss = 12.142976760864258
step = 17800: loss = 28.95887565612793
step = 18000: loss = 334.6312561035156
step = 18000: Average Return = 184.8000030517578
step = 18200: loss = 14.26879596710205
step = 18400: loss = 375.1312561035156
step = 18600: loss = 295.83953857421875
step = 18800: loss = 38.81288528442383
step = 19000: loss = 20.399158477783203
step = 19000: Average Return = 182.5
step = 19200: loss = 33.21414566040039
step = 19400: loss = 32.32746124267578
step = 19600: loss = 706.607177734375
step = 19800: loss = 381.3253479003906
step = 20000: loss = 349.1406555175781
step = 20000: Average Return = 178.89999389648438

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)
(4.125000047683715, 250.0)

png

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.

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.