Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Major code refactoring #6

Merged
merged 7 commits into from
Jan 29, 2018
Merged

Major code refactoring #6

merged 7 commits into from
Jan 29, 2018

Conversation

haarnoja
Copy link
Owner

This pull request replaces large parts of the implementation with soft actor-critic code for better compatibility and easier maintenance. It also changes the way how the action bounds are enforced by replacing InputBounds with a squashing function (tanh).

…parts (#9)

* Replace RLAlgorithm with the version from SAC.
* Replace ReplayBuffer and miscellaneous helper functions.
* Replace NNQFunction and Plotter.
* Replace StochasticNNPolicy.
* Change `Kernel` class into a function.
* Eliminate the need for custom rllab
* Instead of a `SerializableTensor` hack, now using similar serialization
scheme as in SAC.
* Add scripts  to test serialization.
* Add option to train on EC2.
* Remove unused files and code
* Add missing `__init__.py`s
* Rename `algos` -> `algorithms` and `envs` -> `environments`
* Overall clean-up here and there
* Sample a different latent for each sample in batch
* Replace hacky `input_bounds` with proper squashing
* Rename `mlp` -> `feedforward_net` and clean it up
* Simplify `SQL` structure and consolidate notation
* Remove temperature parameter
* Add wrapper for Gym environments
* Update multigoal example.
* Revert multigoal env.
* Clean `Dockerfile` and `environment.yml`
* Update `README.md`
@haarnoja haarnoja merged commit 59c0bbb into master Jan 29, 2018
@haarnoja haarnoja deleted the refactor/master branch January 29, 2018 00:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

1 participant