MatthewGerber/rlai

Issues

• Revisit Swimmer-v4

  #61 opened 23 days ago by MatthewGerber
  1

• Add mypy

  #52 opened 2 months ago by MatthewGerber
  0

• Refactor tests directory layout to match src

  #51 opened 3 months ago by MatthewGerber
  1

• Update github workflows to Python 3.11 and pyproject.toml

  #58 opened 4 months ago by MatthewGerber
  1

• Rerun continuous mountain car

  #57 opened 9 months ago by MatthewGerber
  0

• Ensure that intercepts are only used where appropriate

  #50 opened 9 months ago by MatthewGerber
  0

• Rerun swimming worm

  #56 opened 9 months ago by MatthewGerber
  0

• Tick label doesn't show up in state/reward scatter plot for cartpole run.

  #59 opened 9 months ago by MatthewGerber
  0

• Remove non-stationary feature scaling

  #48 opened 10 months ago by MatthewGerber
  0

• Use state-dimension segments

  #49 opened 10 months ago by MatthewGerber
  0

• Rerun lunar lander after fixing bug in leg contact

  #55 opened 10 months ago by MatthewGerber
  0

• Revive Jupyter Notebook

  #54 opened 10 months ago by MatthewGerber
  0

• Fix robocode tests

  #53 opened 10 months ago by MatthewGerber
  0

• Reconcile td alpha with function approximation step size (currently ignored)

  #15 opened 3 years ago by MatthewGerber
  0

• Get coefficient boxplots to work for actions with varying numbers of features

  #28 opened 3 years ago by MatthewGerber
  0

• Reward shifting should be delegated to agent

  #27 opened 3 years ago by MatthewGerber
  0

• Remove epsilon passed to GPI functions

  #20 opened 4 years ago by MatthewGerber
  0

• Update example dependency project to latest rlai version

  #13 opened 4 years ago by MatthewGerber
  0

• Increase test coverage

  #9 opened 4 years ago by MatthewGerber
  0

• Text/explore passing arguments to scikit-learn SGD model.

  #8 opened 4 years ago by MatthewGerber
  0

• Refactor code to gather up environments, states, and extractors per environment.

  #14 opened 4 years ago by MatthewGerber
  0

• Build up examples of practical use/extension

  #7 opened 4 years ago by MatthewGerber
  0

• Refactor entry point to be rlai with sub-commands for train, agent in environment, etc.

  #11 opened 4 years ago by MatthewGerber
  0

• Chain argument parsers to display help

  #10 opened 4 years ago by MatthewGerber
  0

• Example/test of state-action interaction feature extractor on command line

  #6 opened 4 years ago by MatthewGerber
  0

• Migrate DP solver functionality from agent into trainer.py

  #1 opened 4 years ago by MatthewGerber
  0