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.

Sign up for GitHub

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

Jump to bottom

ActorCritic and DDPG #118

Merged
merged 49 commits into from
Apr 19, 2017
Merged

ActorCritic and DDPG #118

merged 49 commits into from
Apr 19, 2017

Commits on Apr 8, 2017

  1. Working discrete actor critic model

    @lgraesser
    lgraesser committed Apr 8, 2017
    Configuration menu
    Copy the full SHA
    a665bf3 View commit details
    Browse the repository at this point in the history

  2. Uncommenting analyze data

    @lgraesser
    lgraesser committed Apr 8, 2017
    Configuration menu
    Copy the full SHA
    c42d991 View commit details
    Browse the repository at this point in the history

Commits on Apr 9, 2017

  1. style fix, scehdule ac experiment

    @kengz
    kengz committed Apr 9, 2017
    Configuration menu
    Copy the full SHA
    6f76d42 View commit details
    Browse the repository at this point in the history

  2. stylefix

    @kengz
    kengz committed Apr 9, 2017
    Configuration menu
    Copy the full SHA
    2f5c642 View commit details
    Browse the repository at this point in the history

Commits on Apr 10, 2017

  1. Adding AC specs

    @lgraesser
    lgraesser committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    0c4bf3a View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'origin/master' into policy-gradient

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    6041209 View commit details
    Browse the repository at this point in the history

  3. add component locks for ActorCritic and DDPG

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    1be0cb3 View commit details
    Browse the repository at this point in the history

  4. schedule ac on cartpole and pendulum

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    e8aab7c View commit details
    Browse the repository at this point in the history

  5. reorder component locks

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    28d3cc8 View commit details
    Browse the repository at this point in the history

  6. add ac discrete component lock, fix and check all ac specs

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    1565a7a View commit details
    Browse the repository at this point in the history

  7. add variance to pendulum gaussian search, narrow search space

    @kengz
    kengz committed Apr 10, 2017
    Configuration menu
    Copy the full SHA
    e9ad662 View commit details
    Browse the repository at this point in the history

Commits on Apr 11, 2017

  1. add action bounds to env_spec

    @kengz
    kengz committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    c3c4ffb View commit details
    Browse the repository at this point in the history

  2. fix boundedpolicy to auto-bound from env-spec

    @kengz
    kengz committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    7eaed26 View commit details
    Browse the repository at this point in the history

  3. Adding Acrobot specs

    @lgraesser
    lgraesser committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    34b1b58 View commit details
    Browse the repository at this point in the history

  4. Merge branch 'policy-gradient' of https://github.com/kengz/openai_lab … 

    …into policy-gradient
    @lgraesser
    lgraesser committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    9b415c3 View commit details
    Browse the repository at this point in the history

  5. schedule other ac experiments

    @kengz
    kengz committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    3bd227f View commit details
    Browse the repository at this point in the history

  6. Merge branch 'policy-gradient' of https://github.com/kengz/openai_lab … 

    …into policy-gradient
    @kengz
    kengz committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    d11abfd View commit details
    Browse the repository at this point in the history

  7. Fixing mem len param

    @lgraesser
    lgraesser committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    e8e6877 View commit details
    Browse the repository at this point in the history

  8. Merge branch 'policy-gradient' of https://github.com/kengz/openai_lab … 

    …into policy-gradient
    @lgraesser
    lgraesser committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    3d553ee View commit details
    Browse the repository at this point in the history

  9. add ddpg fix attempt

    @kengz
    kengz committed Apr 11, 2017
    Configuration menu
    Copy the full SHA
    75198e4 View commit details
    Browse the repository at this point in the history

Commits on Apr 16, 2017

  1. Merge remote-tracking branch 'origin/master' into policy-gradient

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    edd2276 View commit details
    Browse the repository at this point in the history

  2. ddpg with bounded actions

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    47a5f8c View commit details
    Browse the repository at this point in the history

  3. Merge remote-tracking branch 'origin/master' into policy-gradient

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    ee011b1 View commit details
    Browse the repository at this point in the history

  4. permami broken with reshape to len manually

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    56e7f95 View commit details
    Browse the repository at this point in the history

  5. fixing permami shape one at a time; absolutely disgusting code

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    6215058 View commit details
    Browse the repository at this point in the history

  6. disgusting ddpg hack running

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    434de8b View commit details
    Browse the repository at this point in the history

  7. comment out print

    @kengz
    kengz committed Apr 16, 2017
    Configuration menu
    Copy the full SHA
    4fa1dbc View commit details
    Browse the repository at this point in the history

Commits on Apr 17, 2017

  1. fucking got it, culprit was predicted_q_val shape

    @kengz
    kengz committed Apr 17, 2017
    Configuration menu
    Copy the full SHA
    a691401 View commit details
    Browse the repository at this point in the history

  2. fix permami typo

    @kengz
    kengz committed Apr 17, 2017
    Configuration menu
    Copy the full SHA
    5a827c3 View commit details
    Browse the repository at this point in the history

  3. runnable ddpg2 from permami, still not working yet

    @kengz
    kengz committed Apr 17, 2017
    Configuration menu
    Copy the full SHA
    fd17088 View commit details
    Browse the repository at this point in the history

Commits on Apr 18, 2017

  1. DDPG2 WORKING AT LAST

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    bc0e3f9 View commit details
    Browse the repository at this point in the history

  2. refactor ddpg and rename methods, variables properly

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    ea48d79 View commit details
    Browse the repository at this point in the history

  3. use tf losses; return critic_loss from run

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    366f229 View commit details
    Browse the repository at this point in the history

  4. restore critic_loss

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    1e79a98 View commit details
    Browse the repository at this point in the history

  5. source ddpg main class from dqn; propagate some param settings properly

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    c4d21b8 View commit details
    Browse the repository at this point in the history

  6. add compatible spec

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    0be8381 View commit details
    Browse the repository at this point in the history

  7. externalize select action to policy

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    7e4c28e View commit details
    Browse the repository at this point in the history

  8. refactor noise policies for ddpg

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    402fb71 View commit details
    Browse the repository at this point in the history

  9. separate critic_lr for Critic

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    22c331f View commit details
    Browse the repository at this point in the history

  10. rename base to NoNoisePolicy as proper

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    9ab35e9 View commit details
    Browse the repository at this point in the history

  11. remove DDPGBoundedPolicy, already built in to DDPG

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    fb6059d View commit details
    Browse the repository at this point in the history

  12. remove useless ddpg examples

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    a1524d1 View commit details
    Browse the repository at this point in the history

  13. rename ddpg2 to ddpg

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    da54f53 View commit details
    Browse the repository at this point in the history

  14. stylefix

    @kengz
    kengz committed Apr 18, 2017
    Configuration menu
    Copy the full SHA
    d30fe83 View commit details
    Browse the repository at this point in the history

Commits on Apr 19, 2017

  1. warn instead of break for component lock

    @kengz
    kengz committed Apr 19, 2017
    Configuration menu
    Copy the full SHA
    e469f72 View commit details
    Browse the repository at this point in the history

  2. mute double dqn recompile both models till performance is fixed

    @kengz
    kengz committed Apr 19, 2017
    Configuration menu
    Copy the full SHA
    3b766a9 View commit details
    Browse the repository at this point in the history

  3. fix graph rendering on single trial by mpl backend switching

    @kengz
    kengz committed Apr 19, 2017
    Configuration menu
    Copy the full SHA
    c650b04 View commit details
    Browse the repository at this point in the history

  4. rename noise policies properly

    @kengz
    kengz committed Apr 19, 2017
    Configuration menu
    Copy the full SHA
    be6c2a9 View commit details
    Browse the repository at this point in the history

  5. add ac, ddpg tests

    @kengz
    kengz committed Apr 19, 2017
    Configuration menu
    Copy the full SHA
    0b50a69 View commit details
    Browse the repository at this point in the history