MCPcopy Create free account

hub / github.com/TianhongDai/reinforcement-learning-algorithms / functions

Functions284 in github.com/TianhongDai/reinforcement-learning-algorithms

⨍ Functions 284◇ Types & classes 57
↓ 25 callersMethodmean
(self)
rl_algorithms/dqn_algos/utils.py:48
↓ 25 callersMethodreset
(self)
rl_algorithms/ddpg/utils.py:15
↓ 25 callersMethodstep
(self, action)
rl_utils/logger/bench.py:51
↓ 16 callersMethodclose
(self)
rl_utils/logger/logger.py:352
↓ 10 callersMethodsample
Gradients will and should *not* pass through this operation. See https://github.com/pytorch/pytorch/issues/4620 for discussion.
rl_algorithms/sac/utils.py:35
↓ 9 callersFunctionget_current
()
rl_utils/logger/logger.py:289
↓ 9 callersMethodstep
(self, ac)
rl_utils/env_wrapper/atari_wrapper.py:301
↓ 8 callersMethod__init__
Warp frames to 84x84 as done in the Nature paper and later work. If the environment uses dictionary observations, `dict_space_key` c
rl_utils/env_wrapper/atari_wrapper.py:138
↓ 8 callersFunctionlogkv
Log a value of some diagnostic Call this once for each diagnostic quantity, each iteration If called many times, last value will be used.
rl_utils/logger/logger.py:193
↓ 8 callersMethodrender
(self, mode='human')
rl_utils/env_wrapper/__init__.py:108
↓ 8 callersMethodreset
(self, **kwargs)
rl_utils/env_wrapper/atari_wrapper.py:309
↓ 6 callersMethodclose
(self)
rl_utils/env_wrapper/__init__.py:91
↓ 6 callersFunctionlog
Write the sequence of args, with no separators, to the console and output files (if you've configured an output file).
rl_utils/logger/logger.py:224
↓ 6 callersMethodlog_prob
:param value: some value, x :param pre_tanh_value: arctanh(x) :return:
rl_algorithms/sac/utils.py:25
↓ 5 callersMethod_force
(self)
rl_utils/env_wrapper/atari_wrapper.py:244
↓ 5 callersFunctioncreate_single_env
(args, rank=0)
rl_utils/env_wrapper/create_env.py:13
↓ 5 callersFunctiondumpkvs
Write all of the diagnostics from the current iteration
rl_utils/logger/logger.py:214
↓ 5 callersMethodget_dir
(self)
rl_utils/logger/logger.py:349
↓ 5 callersFunctionmake_atari
(env_id, max_episode_steps=None)
rl_utils/env_wrapper/atari_wrapper.py:269
↓ 5 callersMethodselect_actions
(self, exploration=True, reparameterize=True)
rl_algorithms/sac/utils.py:67
↓ 5 callersFunctionset_seeds
(args, rank=0)
rl_utils/seeds/seeds.py:6
↓ 5 callersFunctionwrap_deepmind
Configure environment for DeepMind-style Atari.
rl_utils/env_wrapper/atari_wrapper.py:278
↓ 4 callersMethod_assert_not_closed
(self)
rl_utils/env_wrapper/multi_envs_wrapper.py:100
↓ 4 callersMethodadd
(self, obs, action, reward, obs_, done)
rl_utils/experience_replay/experience_replay.py:16
↓ 4 callersMethodlog
(self, *args, level=INFO)
rl_utils/logger/logger.py:337
↓ 3 callersMethod__init__
(self, num_envs, observation_space, action_space)
rl_utils/env_wrapper/__init__.py:41
↓ 3 callersMethod_get_tensor_inputs
(self, obs)
rl_algorithms/sac/sac_agent.py:134
↓ 3 callersMethod_get_tensors
(self, obs)
rl_algorithms/a2c/a2c_agent.py:126
↓ 3 callersMethod_get_tensors
(self, obs)
rl_algorithms/ppo/ppo_agent.py:200
↓ 3 callersMethod_get_tensors
(self, obs)
rl_algorithms/dqn_algos/dqn_agent.py:116
↓ 3 callersMethod_mpi_average
(self, x)
rl_utils/mpi_utils/normalizer.py:60
↓ 3 callersFunctionconfigure
If comm is provided, average all numerical stats across that comm
rl_utils/logger/logger.py:372
↓ 3 callersFunctioninfo
(*args)
rl_utils/logger/logger.py:233
↓ 3 callersMethodnormalize
(self, v, clip_range=None)
rl_utils/mpi_utils/normalizer.py:67
↓ 3 callersFunctionplot_results
(dirs, num_timesteps, xaxis, task_name, plt_order, beta=False)
rl_utils/logger/plot.py:67
↓ 3 callersMethodupdate
(self, ob, rew, done, info)
rl_utils/logger/bench.py:58
↓ 2 callersMethod__init__
(self, state_size, num_actions, dist_type)
rl_algorithms/ppo/models.py:11
↓ 2 callersMethod__init__
(self, num_states, num_actions)
rl_algorithms/trpo/models.py:31
↓ 2 callersFunction_flatten_obs
(obs)
rl_utils/env_wrapper/multi_envs_wrapper.py:107
↓ 2 callersFunction_get_flat_params_or_grads
include two kinds: grads and params
rl_utils/mpi_utils/utils.py:25
↓ 2 callersMethod_get_ob
(self)
rl_utils/env_wrapper/atari_wrapper.py:218
↓ 2 callersMethod_get_tensors
(self, obs)
rl_algorithms/trpo/trpo_agent.py:173
↓ 2 callersMethod_preproc_inputs
(self, obs)
rl_algorithms/ddpg/ddpg_agent.py:110
↓ 2 callersFunction_set_flat_params_or_grads
include two kinds: grads and params
rl_utils/mpi_utils/utils.py:33
↓ 2 callersMethod_soft_update_target_network
(self, target, source)
rl_algorithms/ddpg/ddpg_agent.py:163
↓ 2 callersMethod_truncate
(self, s)
rl_utils/logger/logger.py:71
↓ 2 callersMethod_update_target_network
(self, target, source)
rl_algorithms/sac/sac_agent.py:190
↓ 2 callersFunctioncreate_multiple_envs
(args)
rl_utils/env_wrapper/create_env.py:37
↓ 2 callersFunctiondebug
(*args)
rl_utils/logger/logger.py:230
↓ 2 callersFunctiondenormalize
(x, mean, std, clip=10)
rl_algorithms/trpo/demo.py:7
↓ 2 callersFunctiondiscount_with_dones
(rewards, dones, gamma)
rl_algorithms/a2c/utils.py:18
↓ 2 callersFunctioneval_actions
(pi, actions)
rl_algorithms/trpo/utils.py:12
↓ 2 callersFunctionevaluate_actions
(pi, actions, dist_type, env_type)
rl_algorithms/ppo/utils.py:21
↓ 2 callersFunctionget_args
()
rl_algorithms/sac/arguments.py:4
↓ 2 callersFunctionget_args
()
rl_algorithms/ddpg/arguments.py:3
↓ 2 callersFunctionget_args
()
rl_algorithms/a2c/arguments.py:3
↓ 2 callersFunctionget_args
()
rl_algorithms/ppo/arguments.py:3
↓ 2 callersFunctionget_args
()
rl_algorithms/dqn_algos/arguments.py:3
↓ 2 callersFunctionget_args
()
rl_algorithms/trpo/arguments.py:3
↓ 2 callersMethodget_log_prob
(self, actions, pre_tanh_value)
rl_algorithms/sac/utils.py:78
↓ 2 callersFunctionlogkv_mean
The same as logkv(), but if called many times, values averaged.
rl_utils/logger/logger.py:201
↓ 2 callersMethodprocess
(self, obs)
rl_utils/env_wrapper/__init__.py:169
↓ 2 callersMethodreset
(self)
rl_utils/env_wrapper/multi_envs_wrapper.py:77
↓ 2 callersFunctionselect_actions
(pi, deterministic=False)
rl_algorithms/a2c/utils.py:6
↓ 2 callersFunctionset_flat_params_to
(model, flat_params)
rl_algorithms/trpo/utils.py:53
↓ 2 callersMethodstd
(self)
rl_utils/running_filter/running_filter.py:38
↓ 2 callersMethodstep_async
Tell all the environments to start taking a step with the given actions. Call step_wait() to get the results of the step.
rl_utils/env_wrapper/__init__.py:59
↓ 2 callersMethodstep_wait
Wait for the step taken with step_async(). Returns (obs, rews, dones, infos): - obs: an array of observations, or a dict of
rl_utils/env_wrapper/__init__.py:71
↓ 2 callersFunctionsync_grads
(network)
rl_utils/mpi_utils/utils.py:17
↓ 2 callersFunctionsync_networks
netowrk is the network you want to sync
rl_utils/mpi_utils/utils.py:6
↓ 1 callersMethod__init__
(self, input_dims, hidden_size, action_dims=None)
rl_algorithms/sac/models.py:8
↓ 1 callersMethod__init__
(self, obs_dims, action_dims)
rl_algorithms/ddpg/models.py:7
↓ 1 callersMethod__init__
(self, num_actions)
rl_algorithms/a2c/models.py:34
↓ 1 callersMethod__init__
(self, num_actions, use_dueling=False)
rl_algorithms/dqn_algos/models.py:32
↓ 1 callersMethod_adjust_learning_rate
(self, update, num_updates)
rl_algorithms/ppo/ppo_agent.py:211
↓ 1 callersFunction_configure_default_logger
()
rl_utils/logger/logger.py:401
↓ 1 callersFunction_demo
()
rl_utils/logger/logger.py:423
↓ 1 callersMethod_do_log
(self, args)
rl_utils/logger/logger.py:358
↓ 1 callersMethod_encode_sample
(self, idx)
rl_utils/experience_replay/experience_replay.py:26
↓ 1 callersMethod_eval_agent
(self)
rl_algorithms/ddpg/ddpg_agent.py:168
↓ 1 callersMethod_evaluate_agent
(self)
rl_algorithms/sac/sac_agent.py:195
↓ 1 callersMethod_fisher_vector_product
(self, v, obs, pi_old)
rl_algorithms/trpo/trpo_agent.py:151
↓ 1 callersMethod_get_kl
(self, obs, pi_old)
rl_algorithms/trpo/trpo_agent.py:164
↓ 1 callersMethod_get_surrogate_loss
(self, obs, adv, actions, pi_old)
rl_algorithms/trpo/trpo_agent.py:143
↓ 1 callersMethod_initial_exploration
(self, exploration_policy='gaussian')
rl_algorithms/sac/sac_agent.py:110
↓ 1 callersMethod_select_actions
(self, pi)
rl_algorithms/ddpg/ddpg_agent.py:116
↓ 1 callersMethod_update_network
(self)
rl_algorithms/ddpg/ddpg_agent.py:127
↓ 1 callersMethod_update_network
(self, obs, returns, actions)
rl_algorithms/a2c/a2c_agent.py:98
↓ 1 callersMethod_update_network
(self, obs, actions, returns, advantages)
rl_algorithms/ppo/ppo_agent.py:148
↓ 1 callersMethod_update_network
(self, samples)
rl_algorithms/dqn_algos/dqn_agent.py:79
↓ 1 callersMethod_update_network
(self, mb_obs, mb_actions, mb_returns, mb_advs)
rl_algorithms/trpo/trpo_agent.py:95
↓ 1 callersMethod_update_newtork
(self)
rl_algorithms/sac/sac_agent.py:139
↓ 1 callersMethodadd_rewards
(self, reward)
rl_algorithms/dqn_algos/utils.py:31
↓ 1 callersFunctionclear_mpi_env_vars
from mpi4py import MPI will call MPI_Init by default. If the child process has MPI environment variables, MPI will think that the child process
rl_utils/env_wrapper/__init__.py:197
↓ 1 callersMethodclose_extras
Clean up the extra resources, beyond what's in this base class. Only runs when not self.closed.
rl_utils/env_wrapper/__init__.py:84
↓ 1 callersFunctionconjugated_gradient
(fvp, b, update_steps, obs, pi_old, residual_tol=1e-10)
rl_algorithms/trpo/utils.py:18
↓ 1 callersMethoddumpkvs
(self)
rl_utils/logger/logger.py:319
↓ 1 callersFunctionevaluate_actions
(pi, actions)
rl_algorithms/a2c/utils.py:14
↓ 1 callersMethodget_images
Return RGB images from each environment
rl_utils/env_wrapper/__init__.py:111
↓ 1 callersFunctionget_rank_without_mpi_import
()
rl_utils/logger/logger.py:363
next →1–100 of 284, ranked by callers