JohannesHeidecke/irl-benchmark issues and pull requests

#49 - Fix generate_expert_data for FrozenLake

Pull Request - State: closed - Opened by Attler about 5 years ago

#48 - Get >90% test coverage

Issue - State: open - Opened by JohannesHeidecke over 5 years ago

#47 - Make documentation work perfectly

Issue - State: open - Opened by AntonOsika over 5 years ago

#46 - Deploying documentation to github pages

Pull Request - State: closed - Opened by AntonOsika over 5 years ago - 1 comment

#45 - MazeWorld environment and WorldModelWrapper

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago - 2 comments

#44 - Big cleanup

Pull Request - State: closed - Opened by dit7ya over 5 years ago

#43 - Update README.md

Issue - State: closed - Opened by JohannesHeidecke over 5 years ago - 1 comment

#42 - Host our docs on readthedocs.org

Issue - State: closed - Opened by JohannesHeidecke over 5 years ago

#41 - ILE calculated using initial/estimated rather than true reward

Issue - State: open - Opened by maxmdaniel over 5 years ago - 1 comment

#40 - RelEnt uses trajectories of varying length

Issue - State: open - Opened by maxmdaniel over 5 years ago - 1 comment

#39 - Inconsistent use of discount factors

Issue - State: open - Opened by maxmdaniel over 5 years ago

#38 - MaxEnt estimates the initial state distribution rather than getting it from the env

Issue - State: open - Opened by maxmdaniel over 5 years ago

#37 - Added plot util

Pull Request - State: closed - Opened by maxmdaniel over 5 years ago

#36 - Experiment script

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#35 - Value iteration and main script

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#34 - Antonosika/mce irl

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#33 - fixing get_reward_function

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#32 - fix pytest

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#31 - Bugfix get reward

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#30 - Fast test, see #28 for more details

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#29 - Implementation of Relative Entropy IRL for rewards linear in features

Pull Request - State: closed - Opened by maxmdaniel over 5 years ago

#28 - Fast tests

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#27 - Maximum causal entropy and GoodValueIteration and Inverse Learning Error

Pull Request - State: closed - Opened by dit7ya over 5 years ago

#26 - Value Iteration uses leaked true reward

Issue - State: closed - Opened by JohannesHeidecke over 5 years ago

#25 - Starting fixes of PPO tests

Pull Request - State: closed - Opened by rhaps0dy over 5 years ago - 2 comments

#24 - Test `tests/ppo_test.py` fails

Issue - State: closed - Opened by rhaps0dy over 5 years ago - 1 comment

#23 - Added neural-network based potential L2 metric

Pull Request - State: closed - Opened by rhaps0dy over 5 years ago - 1 comment

#22 - config file mapping from IRL-algorithms to usable environments

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#21 - make function for FeatureWrappers

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#20 - Add comet examples and MetricsListener

Pull Request - State: closed - Opened by AntonOsika over 5 years ago - 2 comments

#19 - L2 minimisation metric working for tabular State cases

Pull Request - State: closed - Opened by rhaps0dy over 5 years ago - 1 comment

#18 - MaxEnt IRL

Pull Request - State: closed - Opened by dit7ya over 5 years ago - 1 comment

#17 - PPO

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#16 - Value iteration and lower duration for q learner test

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#15 - unwrap_env and get_transition_matrix helpers

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#14 - Sayan fresh

Pull Request - State: closed - Opened by dit7ya over 5 years ago

#13 - Test save loading and use pytest

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#12 - Add policy to rl algorithm and clarify readme

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#11 - added remaining docstrings

Pull Request - State: closed - Opened by maxmdaniel over 5 years ago - 1 comment

#10 - Add rl alg factory to irl base [MERGE PREVIOUS PR FIRST]

Pull Request - State: closed - Opened by AntonOsika over 5 years ago - 1 comment

#9 - Rename FrozenFeatureWrapper (Frozen = reserved name) to FrozenLake...

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#8 - small cleanup & adding some true reward functions

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#7 - Added docstrings to examples, algorithms, metrics.

Pull Request - State: closed - Opened by maxmdaniel over 5 years ago

#6 - Python package

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#5 - Max/projection

Pull Request - State: closed - Opened by maxmdaniel over 5 years ago

#4 - Antonosika/python package [DO NOT MERGE]

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#3 - Johannes

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago

#2 - Antonosika/tabular q

Pull Request - State: closed - Opened by AntonOsika over 5 years ago

#1 - created feature wrapper and feature-based reward functions

Pull Request - State: closed - Opened by JohannesHeidecke over 5 years ago - 2 comments

GitHub / JohannesHeidecke/irl-benchmark issues and pull requests