* Update documentation
Added comment to PPO documentation that CPU should primarily be used unless using CNN as well as sample code. Added warning to user for both PPO and A2C that CPU should be used if the user is running GPU without using a CNN, reference Issue #1245.
* Add warning to base class and add test
---------
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Fix failing set_env test
* Fix test failiing due to deprectation of env.seed
* Adjust mean reward threshold in failing test
* Fix her test failing due to rng
* Change seed and revert reward threshold to 90
* Pin gym version
* Make VecEnv compatible with gym seeding change
* Revert change to VecEnv reset signature
* Change subprocenv seed cmd to call reset instead
* Fix type check
* Add backward compat
* Add `compat_gym_seed` helper
* Add goal env checks in env_checker
* Add docs on HER requirements for envs
* Capture user warning in test with inverted box space
* Update ale-py version
* Fix randint
* Allow noop_max to be zero
* Update changelog
* Update docker image
* Update doc conda env and dockerfile
* Custom envs should not have any warnings
* Fix test for numpy >= 1.21
* Add check for vectorized compute reward
* Bump to gym 0.24
* Fix gym default step docstring
* Test downgrading gym
* Revert "Test downgrading gym"
This reverts commit 0072b77156c006ada8a1d6e26ce347ed85a83eeb.
* Fix protobuf error
* Fix in dependencies
* Fix protobuf dep
* Use newest version of cartpole
* Update gym
* Fix warning
* Loosen required scipy version
* Scipy no longer needed
* Try gym 0.25
* Silence warnings from gym
* Filter warnings during tests
* Update doc
* Update requirements
* Add gym 26 compat in vec env
* Fixes in envs and tests for gym 0.26+
* Enforce gym 0.26 api
* format
* Fix formatting
* Fix dependencies
* Fix syntax
* Cleanup doc and warnings
* Faster tests
* Higher budget for HER perf test (revert prev change)
* Fixes and update doc
* Fix doc build
* Fix breaking change
* Fixes for rendering
* Rename variables in monitor
* update render method for gym 0.26 API
backwards compatible (mode argument is allowed) while using the gym 0.26 API (render mode is determined at environment creation)
* update tests and docs to new gym render API
* undo removal of render modes metatadata check
* set rgb_array as default render mode for gym.make
* undo changes & raise warning if not 'rgb_array'
* Fix type check
* Remove recursion and fix type checking
* Remove hacks for protobuf and gym 0.24
* Fix type annotations
* reuse existing render_mode attribute
* return tiled images for 'human' render mode
* Allow to use opencv for human render, fix typos
* Add warning when using non-zero start with Discrete (fixes#1197)
* Fix type checking
* Bug fixes and handle more cases
* Throw proper warnings
* Update test
* Fix new metadata name
* Ignore numpy warnings
* Fixes in vec recorder
* Global ignore
* Filter local warning too
* Monkey patch not needed for gym 26
* Add doc of VecEnv vs Gym API
* Add render test
* Fix return type
* Update VecEnv vs Gym API doc
* Fix for custom render mode
* Fix return type
* Fix type checking
* check test env test_buffer
* skip render check
* check env test_dict_env
* test_env test_gae
* check envs in remaining tests
* Update tests
* Add warning for Discrete action space with non-zero (#1295)
* Fix atari annotation
* ignore get_action_meanings [attr-defined]
* Fix mypy issues
* Add patch for gym/gymnasium transition
* Switch to gymnasium
* Rely on signature instead of version
* More patches
* Type ignore because of https://github.com/Farama-Foundation/Gymnasium/pull/39
* Fix doc build
* Fix pytype errors
* Fix atari requirement
* Update env checker due to change in dtype for Discrete
* Fix type hint
* Convert spaces for saved models
* Ignore pytype
* Remove gitlab CI
* Disable pytype for convert space
* Fix undefined info
* Fix undefined info
* Upgrade shimmy
* Fix wrappers type annotation (need PR from Gymnasium)
* Fix gymnasium dependency
* Fix dependency declaration
* Cap pygame version for python 3.7
* Point to master branch (v0.28.0)
* Fix: use main not master branch
* Rename done to terminated
* Fix pygame dependency for python 3.7
* Rename gym to gymnasium
* Update Gymnasium
* Fix test
* Fix tests
* Forks don't have access to private variables
* Fix linter warnings
* Update read the doc env
* Fix env checker for GoalEnv
* Fix import
* Update env checker (more info) and fix dtype
* Use micromamab for Docker
* Update dependencies
* Clarify VecEnv doc
* Fix Gymnasium version
* Copy file only after mamba install
* [ci skip] Update docker doc
* Polish code
* Reformat
* Remove deprecated features
* Ignore warning
* Update doc
* Update examples and changelog
* Fix type annotation bundle (SAC, TD3, A2C, PPO, base class) (#1436)
* Fix SAC type hints, improve DQN ones
* Fix A2C and TD3 type hints
* Fix PPO type hints
* Fix on-policy type hints
* Fix base class type annotation, do not use defaults
* Update version
* Disable mypy for python 3.7
* Rename Gym26StepReturn
* Update continuous critic type annotation
* Fix pytype complain
---------
Co-authored-by: Carlos Luis <carlos.luisgonc@gmail.com>
Co-authored-by: Quentin Gallouédec <45557362+qgallouedec@users.noreply.github.com>
Co-authored-by: Thomas Lips <37955681+tlpss@users.noreply.github.com>
Co-authored-by: tlips <thomas.lips@ugent.be>
Co-authored-by: tlpss <thomas17.lips@gmail.com>
Co-authored-by: Quentin GALLOUÉDEC <gallouedec.quentin@gmail.com>
* fix Atari in CI
* fix dtype and atari extra
* Update setup.py
* remove 3.6
* note about how to install Atari
* pendulum-v1
* atari v5
* black
* fix pendulum capitalization
* add minimum version
* moved things in changelog to breaking changes
* partial v5 fix
* env update to pass tests
* mismatch env version fixed
* Fix tests after merge
* Include autorom in setup.py
* Blacken code
* Fix dtype issue in more robust way
* Fix GitLab CI: switch to Docker container with new black version
* Remove workaround from GitLab. (May need to rebuild Docker for this though.)
* Revert to v4
* Update setup.py
* Apply suggestions from code review
* Remove unnecessary autorom
* Consistent gym versions
Co-authored-by: J K Terry <justinkterry@gmail.com>
Co-authored-by: Anssi <kaneran21@hotmail.com>
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
Co-authored-by: modanesh <mohamad4danesh@gmail.com>
Co-authored-by: Adam Gleave <adam@gleave.me>
* Add multi-env training support for SAC
* Fix for dict obs
* Pytype fixes
* Fix assert on number of envs
* Remove for loop
* Add support for Dict obs
* Start cleanup
* Update doc and bug fix
* Add support for vectorized action noise
and add multi env example for off-policy
* Update version
* Bug fix with VecNormalize
* Update README table
* Update variable names
* Update changelog and version
* Update doc and fix for `gradient_steps=-1`
* Add test for `gradient_steps=-1`
* Disable pytype pyi errors
* Fix for DQN
* Update comment on deepcopy
* Remove episode_reward field
* Fix RolloutReturn
* Avoid modification by reference
* Fix error message
Co-authored-by: Anssi <kaneran21@hotmail.com>
* Add support for custom objects
* Add python 3.8 to the CI
* Bump version
* PyType fixes
* [ci skip] Fix typo
* Add note about slow-down + fix typos
* Minor edits to the doc
* Bug fix for DQN
* Update test
* Add test for custom objects
* Removed unneeded overrides of feature_extractor and normalize_images in the TD3 Actor.
* Add learning rate schedule example (#248)
* Add learning rate schedule example
* Update docs/guide/examples.rst
Co-authored-by: Adam Gleave <adam@gleave.me>
* Address comments
Co-authored-by: Adam Gleave <adam@gleave.me>
* Add supported action spaces checks (#254)
* Add supported action spaces checks
* Address comment
* Use `pass` in an abstractmethod instead of deleting the arguments.
* Remove the "deterministic" keyword from the forward method of the TD3 Actor since it always is deterministic anyways.
* Rename _get_data to _get_data_to_reconstruct_model.
_get_data was too generic and could have meant anything.
* Remove the n_episodes_rollout parameter and allow passing tuples as train_freq instead.
* Fix docstring of `train_freq` parameter.
* Black fixes.
* Fix TD3 delayed update + rename `_get_data()`
* Fix TD3 test
* Normalize `train_freq` to a tuple in the constructor and turn the warning into an assert.
* Make one step the default train frequency.
* Black fixes.
* Change np.bool to bool.
* Use the tuple format to specify an amount of steps in terms of steps or episodes in the collect_collouts of the off policy algorithm.
* Use the tuple format to specify an amount of steps in terms of steps or episodes in the collect_collouts of HER.
* Use named tuple for train freq
* Rename train_freq to train_every and TrainFreq to ExperienceDuration. Also add some type annotations and documentation.
* Black fixes.
* Revert to train_freq
* Fix terminal observation issues
* Typo
* Fix action noise bug in HER
* Add assert when loading HER models
* Update version
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
Co-authored-by: Adam Gleave <adam@gleave.me>
* Update doc and add new example
* Add save/load replay buffer example
* Add save format + export doc
* Add example for get/set parameters
* Typos and minor edits
* Add results sections
* Add note about performance
* Add DDPG results
* Address comments
* Fix grammar/wording
Co-authored-by: Anssi "Miffyli" Kanervisto <kaneran21@hotmail.com>
* Added working her version, Online sampling is missing.
* Updated test_her.
* Added first version of online her sampling. Still problems with tensor dimensions.
* Reformat
* Fixed tests
* Added some comments.
* Updated changelog.
* Add missing init file
* Fixed some small bugs.
* Reduced arguments for HER, small changes.
* Added getattr. Fixed bug for online sampling.
* Updated save/load funtions. Small changes.
* Added her to init.
* Updated save method.
* Updated her ratio.
* Move obs_wrapper
* Added DQN test.
* Fix potential bug
* Offline and online her share same sample_goal function.
* Changed lists into arrays.
* Updated her test.
* Fix online sampling
* Fixed action bug. Updated time limit for episodes.
* Updated convert_dict method to take keys as arguments.
* Renamed obs dict wrapper.
* Seed bit flipping env
* Remove get_episode_dict
* Add fast online sampling version
* Added documentation.
* Vectorized reward computation
* Vectorized goal sampling
* Update time limit for episodes in online her sampling.
* Fix max episode length inference
* Bug fix for Fetch envs
* Fix for HER + gSDE
* Reformat (new black version)
* Added info dict to compute new reward. Check her_replay_buffer again.
* Fix info buffer
* Updated done flag.
* Fixes for gSDE
* Offline her version uses now HerReplayBuffer as episode storage.
* Fix num_timesteps computation
* Fix get torch params
* Vectorized version for offline sampling.
* Modified offline her sampling to use sample method of her_replay_buffer
* Updated HER tests.
* Updated documentation
* Cleanup docstrings
* Updated to review comments
* Fix pytype
* Update according to review comments.
* Removed random goal strategy. Updated sample transitions.
* Updated migration. Removed time signal removal.
* Update doc
* Fix potential load issue
* Add VecNormalize support for dict obs
* Updated saving/loading replay buffer for HER.
* Fix test memory usage
* Fixed save/load replay buffer.
* Fixed save/load replay buffer
* Fixed transition index after loading replay buffer in online sampling
* Better error handling
* Add tests for get_time_limit
* More tests for VecNormalize with dict obs
* Update doc
* Improve HER description
* Add test for sde support
* Add comments
* Add comments
* Remove check that was always valid
* Fix for terminal observation
* Updated buffer size in offline version and reset of HER buffer
* Reformat
* Update doc
* Remove np.empty + add doc
* Fix loading
* Updated loading replay buffer
* Separate online and offline sampling + bug fixes
* Update tensorboard log name
* Version bump
* Bug fix for special case
Co-authored-by: Antonin Raffin <antonin.raffin@dlr.de>
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Fix storing correct episode dones
* Fix number of filters in NatureCNN network
* Add TF-like RMSprop for matching performance with sb2
* Remove stuff that was accidentally included
* Reformat
* Clarify variable naming
* Update changelog
* Add comment on RMSprop implementations to A2C
* Add test for RMSpropTFLike
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Add DDPG + TD3 with any number of critics
* Allow any number of critics for SAC
* Update doc
* [ci skip] Update DDPG example
* Remove unused parameter
* Add DDPG to identity test
* Fix computation with n_critics=1,3
* Update doc
* Apply suggestions from code review
Co-authored-by: Adam Gleave <adam@gleave.me>
* Update docstrings for off-policy algos
* Add check for sde
Co-authored-by: Adam Gleave <adam@gleave.me>
* Created DQN template according to the paper.
Next steps:
- Create Policy
- Complete Training
- Debug
* Changed Base Class
* refactor save, to be consistence with overriding the excluded_save_params function. Do not try to exclude the parameters twice.
* Added simple DQN policy
* Finished learn and train function
- missing correct loss computation
* changed collect_rollouts to work with discrete space
* moved discrete space collect_rollouts to dqn
* basic dqn working
* deleted SDE related code
* added gradient clipping and moved greedy policy to policy
* changed policy to implement target network
and added soft update(in fact standart tau is 1 so hard update)
* fixed policy setup
* rebase target_update_intervall on _n_updates
* adapted all tests
all tests passing
* Move to stable-baseline3
* Fixes for DQN
* Fix tests + add CNNPolicy
* Allow any optimizer for DQN
* added some util functions to create a arbitrary linear schedule, fixed pickle problem with old exploration schedule
* more documentation
* changed buffer dtype
* refactor and document
* Added Sphinx Documentation
Updated changelog.rst
* removed custom collect_rollouts as it is no longer necessary
* Implemented suggestions to clean code and documentation.
* extracted some functions on tests to reduce duplicated code
* added support for exploration_fraction
* Fixed exploration_fraction
* Added documentation
* Fixed get_linear_fn -> proper progress scaling
* Merged master
* Added nature reference
* Changed default parameters to https://www.nature.com/articles/nature14236/tables/1
* Fixed n_updates to be incremented correctly
* Correct train_freq
* Doc update
* added special parameter for DQN in tests
* different fix for test_discrete
* Update docs/modules/dqn.rst
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Update docs/modules/dqn.rst
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Update docs/modules/dqn.rst
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Added RMSProp in optimizer_kwargs, as described in nature paper
* Exploration fraction is inverse of 50.000.000 (total frames) / 1.000.000 (frames with linear schedule) according to nature paper
* Changelog update for buffer dtype
* standard exlude parameters should be always excluded to assure proper saving only if intentionally included by ``include`` parameter
* slightly more iterations on test_discrete to pass the test
* added param use_rms_prop instead of mutable default argument
* forgot alpha
* using huber loss, adam and learning rate 1e-4
* account for train_freq in update_target_network
* Added memory check for both buffers
* Doc updated for buffer allocation
* Added psutil Requirement
* Adapted test_identity.py
* Fixes with new SB3 version
* Fix for tensorboard name
* Convert assert to warning and fix tests
* Refactor off-policy algorithms
* Fixes
* test: remove next_obs in replay buffer
* Update changelog
* Fix tests and use tmp_path where possible
* Fix sampling bug in buffer
* Do not store next obs on episode termination
* Fix replay buffer sampling
* Update comment
* moved epsilon from policy to model
* Update predict method
* Update atari wrappers to match SB2
* Minor edit in the buffers
* Update changelog
* Merge branch 'master' into dqn
* Update DQN to new structure
* Fix tests and remove hardcoded path
* Fix for DQN
* Disable memory efficient replay buffer by default
* Fix docstring
* Add tests for memory efficient buffer
* Update changelog
* Split collect rollout
* Move target update outside `train()` for DQN
* Update changelog
* Update linear schedule doc
* Cleanup DQN code
* Minor edit
* Update version and docker images
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
* Split torch module code into torch_layers file
* Updated reference to CNN
* Change 'CxWxH' to 'CxHxW', as per common notion
* Fix missing import in policies.py
* Move PPOPolicy to OnlineActorCriticPolicy
* Create OnPolicyRLModel from PPO, and make A2C and PPO inherit
* Update A2C optimizer comment
* Clean weight init scales for clarity
* Fix A2C log_interval default parameter
* Rename 'progress' to 'progress_remaining
* Rename 'Models' to 'Algorithms'
* Rename 'OnlineActorCriticPolicy' to 'ActorCriticPolicy'
* Move static functions out from BaseAlgorithm
* Move on/off_policy base algorithms to their own files
* Add files for A2C/PPO
* Fix docs
* Fix pytype
* Update documentation on OnPolicyAlgorithm
* Add proper doctstring for on_policy rollout gathering
* Add bit clarification on the mlppolicy/cnnpolicy naming
* Move static function is_vectorized_policies to utils.py
* Checking docstrings, pep8 fixes
* Update changelog
* Clean changelog
* Remove policy warnings for sac/td3
* Add monitor_wrapper for OnPolicyAlgorithm. Clean tb logging variables. Add parameter keywords to OffPolicyAlgorithm super init
Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
