def test_fit_batch_online_atari_with_dqn():
import d4rl_atari
make_env = lambda: ChannelFirst(DummyAtari())
env = AsyncBatchEnv([make_env for _ in range(2)])
eval_env = ChannelFirst(DummyAtari())
algo = DQN(n_frames=4)
buffer = BatchReplayBuffer(1000, env)
explorer = LinearDecayEpsilonGreedy()
algo.fit_batch_online(
env,
buffer,
explorer,
n_epochs=1,
n_steps_per_epoch=500,
n_updates_per_epoch=1,
eval_env=eval_env,
logdir="test_data",
)
assert algo.impl.observation_shape == (4, 84, 84)
def test_fit_batch_online_atari_with_dqn():
import d4rl_atari
make_env = lambda: gym.make("breakout-mixed-v0", stack=False)
env = AsyncBatchEnv([make_env for _ in range(2)])
eval_env = gym.make("breakout-mixed-v0", stack=False)
algo = DQN(n_frames=4)
buffer = BatchReplayBuffer(1000, env)
explorer = LinearDecayEpsilonGreedy()
algo.fit_batch_online(
env,
buffer,
explorer,
n_epochs=1,
n_steps_per_epoch=500,
n_updates_per_epoch=1,
eval_env=eval_env,
logdir="test_data",
tensorboard=False,
)
assert algo.impl.observation_shape == (4, 84, 84)
def test_async_batch_env_continuous(n_envs, n_steps):
make_env_fn = lambda: gym.make("Pendulum-v0")
env = AsyncBatchEnv([make_env_fn for _ in range(n_envs)])
observation_shape = env.observation_space.shape
action_size = env.action_space.shape[0]
observations = env.reset()
assert observations.shape == (n_envs, ) + observation_shape
for _ in range(n_steps):
actions = np.random.random((n_envs, action_size))
observations, rewards, terminals, infos = env.step(actions)
assert observations.shape == (n_envs, ) + observation_shape
assert rewards.shape == (n_envs, )
assert terminals.shape == (n_envs, )
assert len(infos) == n_envs
def test_async_batch_env_discrete(n_envs, n_steps):
make_env_fn = lambda: gym.make("CartPole-v0")
env = AsyncBatchEnv([make_env_fn for _ in range(n_envs)])
observation_shape = env.observation_space.shape
action_size = env.action_space.n
observations = env.reset()
assert observations.shape == (n_envs, ) + observation_shape
for _ in range(n_steps):
actions = np.random.randint(action_size, size=n_envs)
observations, rewards, terminals, infos = env.step(actions)
assert observations.shape == (n_envs, ) + observation_shape
assert rewards.shape == (n_envs, )
assert terminals.shape == (n_envs, )
assert len(infos) == n_envs
def test_fit_batch_online_pendulum_with_sac():
make_env = lambda: gym.make("Pendulum-v0")
env = AsyncBatchEnv([make_env for _ in range(5)])
eval_env = gym.make("Pendulum-v0")
algo = SAC()
buffer = BatchReplayBuffer(1000, env)
algo.fit_batch_online(
env,
buffer,
n_epochs=1,
n_steps_per_epoch=500,
n_updates_per_epoch=1,
eval_env=eval_env,
logdir="test_data",
)
def test_fit_batch_online_cartpole_with_dqn():
make_env = lambda: gym.make("CartPole-v0")
env = AsyncBatchEnv([make_env for _ in range(5)])
eval_env = gym.make("CartPole-v0")
algo = DQN()
buffer = BatchReplayBuffer(1000, env)
explorer = LinearDecayEpsilonGreedy()
algo.fit_batch_online(
env,
buffer,
explorer,
n_epochs=1,
n_steps_per_epoch=500,
n_updates_per_epoch=1,
eval_env=eval_env,
logdir="test_data",
)
import gym
from d3rlpy.algos import DQN
from d3rlpy.envs import AsyncBatchEnv
from d3rlpy.online.buffers import BatchReplayBuffer
from d3rlpy.online.explorers import LinearDecayEpsilonGreedy
if __name__ == '__main__':
env = AsyncBatchEnv([lambda: gym.make('CartPole-v0') for _ in range(10)])
eval_env = gym.make('CartPole-v0')
# setup algorithm
dqn = DQN(batch_size=32,
learning_rate=1e-3,
target_update_interval=1000,
use_gpu=False)
# replay buffer for experience replay
buffer = BatchReplayBuffer(maxlen=100000, env=env)
# epilon-greedy explorer
explorer = LinearDecayEpsilonGreedy(start_epsilon=1.0,
end_epsilon=0.1,
duration=100000)
# start training
dqn.fit_batch_online(env,
buffer,
explorer,
n_epochs=100,
eval_interval=1,
import gym
from d3rlpy.algos import SAC
from d3rlpy.envs import AsyncBatchEnv
from d3rlpy.online.buffers import BatchReplayBuffer
if __name__ == '__main__':
env = AsyncBatchEnv([lambda: gym.make('Pendulum-v0') for _ in range(10)])
eval_env = gym.make('Pendulum-v0')
# setup algorithm
sac = SAC(batch_size=100, use_gpu=False)
# replay buffer for experience replay
buffer = BatchReplayBuffer(maxlen=100000, env=env)
# start training
sac.fit_batch_online(env,
buffer,
n_epochs=100,
eval_interval=1,
eval_env=eval_env,
n_steps_per_epoch=1000,
n_updates_per_epoch=1000)
