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

[RL-baseline] Model v5 #42

Open
wants to merge 28 commits into
base: RL-with-baseline
Choose a base branch
from
Open

[RL-baseline] Model v5 #42

wants to merge 28 commits into from

Commits on Mar 11, 2021

  1. Add metrics and logsoftmax

    @xeviknal
    xeviknal committed Mar 11, 2021
    Configuration menu
    Copy the full SHA
    2106b4b View commit details
    Browse the repository at this point in the history

  2. Updating the model

    @xeviknal
    xeviknal committed Mar 11, 2021
    Configuration menu
    Copy the full SHA
    0f2f195 View commit details
    Browse the repository at this point in the history

  3. Add new model to baseline

    @xeviknal
    xeviknal committed Mar 11, 2021
    Configuration menu
    Copy the full SHA
    39b3ad3 View commit details
    Browse the repository at this point in the history

Commits on Mar 14, 2021

  1. The line that fixes all

    @xeviknal @ziritrion
    xeviknal authored and ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    907af7a View commit details
    Browse the repository at this point in the history

  2. Add mean entropy - to reduce tensorboard runs

    @xeviknal @ziritrion
    xeviknal authored and ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    8e4ee6c View commit details
    Browse the repository at this point in the history

  3. Add action prob mean: mean of prob of actions taken in the episode

    @xeviknal
    xeviknal committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    3970702 View commit details
    Browse the repository at this point in the history

  4. Added simple directory check to params folder

    @ziritrion
    ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    957a3b4 View commit details
    Browse the repository at this point in the history

  5. Added additional param save conditions (end of log_interval, last epi… 

    …sode, successful training)
    @ziritrion
    ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    0105564 View commit details
    Browse the repository at this point in the history

  6. Merge branch 'RL-baseline-new-model' of github.com:xeviknal/aidl-2021… 

    …-wo-rl into RL-baseline-new-model

Trying to solve conflicts
    @ziritrion
    ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    b5a5184 View commit details
    Browse the repository at this point in the history

  7. Removing old runs; they don't apply to this branch

    @ziritrion
    ziritrion committed Mar 14, 2021
    Configuration menu
    Copy the full SHA
    c6954ec View commit details
    Browse the repository at this point in the history

Commits on Mar 15, 2021

  1. RL-baseline-NM-save-optim

    @xeviknal @ziritrion
    xeviknal authored and ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    a7c907c View commit details
    Browse the repository at this point in the history

  2. Load optimizer params

    @xeviknal @ziritrion
    xeviknal authored and ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    d5b676c View commit details
    Browse the repository at this point in the history

  3. 8k runs

    @ziritrion
    ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    bd7f6c0 View commit details
    Browse the repository at this point in the history

  4. Fresh start with latest checkpoint load-save changes. Also, small git… 

    …ignore change
    @ziritrion
    ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    5f246c0 View commit details
    Browse the repository at this point in the history

  5. bugfix

    @ziritrion
    ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    e8aa5e4 View commit details
    Browse the repository at this point in the history

  6. 10k runs

    @ziritrion
    ziritrion committed Mar 15, 2021
    Configuration menu
    Copy the full SHA
    c52a4f2 View commit details
    Browse the repository at this point in the history

Commits on Mar 16, 2021

  1. Almost 20k runs. Reward is starting to improve little by little

    @ziritrion
    ziritrion committed Mar 16, 2021
    Configuration menu
    Copy the full SHA
    192bece View commit details
    Browse the repository at this point in the history

  2. Fixed runner.py for generating videos

    @ziritrion
    ziritrion committed Mar 16, 2021
    Configuration menu
    Copy the full SHA
    11b46c4 View commit details
    Browse the repository at this point in the history

  3. 25k runs. Slight improvement but far from desirable

    @ziritrion
    ziritrion committed Mar 16, 2021
    Configuration menu
    Copy the full SHA
    befd201 View commit details
    Browse the repository at this point in the history

Commits on Mar 23, 2021

  1. Adding more linear layers for each head

    @xeviknal
    xeviknal committed Mar 23, 2021
    Configuration menu
    Copy the full SHA
    610d5a8 View commit details
    Browse the repository at this point in the history

Commits on Mar 25, 2021

  1. Cleaned up the code and removed params and runs in order to ease merg… 

    …ing with other branches
    @ziritrion
    ziritrion committed Mar 25, 2021
    Configuration menu
    Copy the full SHA
    e9ad814 View commit details
    Browse the repository at this point in the history

  2. Modified actions to include the no_action possibility, as well as fin… 

    …etuning
    @ziritrion
    ziritrion committed Mar 25, 2021
    Configuration menu
    Copy the full SHA
    4409cf6 View commit details
    Browse the repository at this point in the history

  3. Tweaked the way the params filename is generated, as well as tensorbo… 

    …ard references for ease of use
    @ziritrion
    ziritrion committed Mar 25, 2021
    Configuration menu
    Copy the full SHA
    db1bd06 View commit details
    Browse the repository at this point in the history

  4. In train_episode, moved some vars to GPU that weren't being moved before

    @ziritrion
    ziritrion committed Mar 25, 2021
    Configuration menu
    Copy the full SHA
    e5a9707 View commit details
    Browse the repository at this point in the history

Commits on Mar 29, 2021

  1. Base start for Baseline model v4

    @ziritrion
    ziritrion committed Mar 29, 2021
    Configuration menu
    Copy the full SHA
    c3aa2c7 View commit details
    Browse the repository at this point in the history

Commits on Mar 30, 2021

  1. Base commit for v5

    @ziritrion
    ziritrion committed Mar 30, 2021
    Configuration menu
    Copy the full SHA
    81ea966 View commit details
    Browse the repository at this point in the history

Commits on Mar 31, 2021

  1. Model v5 finalized

    @ziritrion
    ziritrion committed Mar 31, 2021
    Configuration menu
    Copy the full SHA
    77db903 View commit details
    Browse the repository at this point in the history

Commits on Apr 3, 2021

  1. Added additional action set with greater granularity

    @ziritrion
    ziritrion committed Apr 3, 2021
    Configuration menu
    Copy the full SHA
    5edc9ca View commit details
    Browse the repository at this point in the history