random_agent.py

import imageio
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.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.specs import tensor_spec
from tf_agents.utils import common

import env

num_iterations = 400 #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"}

env_name = 'games/SpaceInvaders2'
env = suite_gym.load(env_name, gym_kwargs={'render_mode': 'rgb_array'})
env.reset()

train_py_env = suite_gym.load(env_name, gym_kwargs={'render_mode': 'rgb_array'})
eval_py_env = suite_gym.load(env_name, gym_kwargs={'render_mode': 'rgb_array'})

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

#fc_layer_params = (100, 50)
fc_layer_params = (32400, 1000, 50)
action_tensor_spec = tensor_spec.from_spec(env.action_spec())
num_actions = action_tensor_spec.maximum - action_tensor_spec.minimum + 1

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])

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()

eval_policy = agent.policy
collect_policy = agent.collect_policy

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

example_environment = tf_py_environment.TFPyEnvironment(
    suite_gym.load(env_name, gym_kwargs={'render_mode': 'rgb_array'}))

time_step = example_environment.reset()

random_policy.action(time_step)

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]


compute_avg_return(eval_env, random_policy, num_eval_episodes)

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)

# agent.collect_data_spec

# agent.collect_data_spec._fields

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())

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

# 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)

iterator = iter(dataset)
print(iterator)

# 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()

# (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)

iterations = range(0, num_iterations + 1, eval_interval)


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().reshape((720, 720, 3)))
      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().reshape((720, 720, 3)))

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