GitHub / google-research/batch_rl issues and pull requests
#41 - Could you please release Score_{min} and Score_{max} for atari games?
Issue -
State: open - Opened by wtrvegeword about 1 year ago
#40 - Is there some errors with the actions in the Ms-Pacman dataset?
Issue -
State: open - Opened by hutchinsonian about 1 year ago
#39 - Windows: basic test failed
Issue -
State: open - Opened by yceny over 1 year ago
#38 - About terminal in Atari dataset
Issue -
State: open - Opened by tatekennn almost 2 years ago
- 1 comment
#37 - configurable() got an unexpected keyword argument 'blacklist'
Issue -
State: closed - Opened by massooti about 2 years ago
- 2 comments
#36 - Update README.md to include a warning on ROM versions
Pull Request -
State: closed - Opened by zhixuan-lin about 2 years ago
- 1 comment
#35 - only action checkpoint in https://console.cloud.google.com/storage/browser/atari-replay-datasets
Issue -
State: closed - Opened by histmeisah over 2 years ago
- 1 comment
#34 - A copy of the dataset?
Issue -
State: closed - Opened by vmoens over 2 years ago
- 1 comment
#33 - Difference b/w checkpoint 49 and checkpoint 50
Issue -
State: open - Opened by etaoxing over 2 years ago
- 5 comments
#32 - About DQN replay dataset
Issue -
State: closed - Opened by sglucas over 2 years ago
- 2 comments
#31 - How to get expert data ?
Issue -
State: closed - Opened by prajjwal1 almost 3 years ago
- 5 comments
#30 - JAX code
Issue -
State: open - Opened by lucasliunju almost 3 years ago
- 11 comments
Labels: enhancement, good first issue
#29 - WARNING:absl:Unable to find episode_end_indices. This is expected for old checkpoints.
Issue -
State: closed - Opened by MinkyungPark over 3 years ago
- 1 comment
#28 - Asterix/1 dataset broken?
Issue -
State: open - Opened by dssrgu over 3 years ago
- 3 comments
#27 - Transform matrix create only once?
Issue -
State: closed - Opened by Hyperion-shuo over 3 years ago
- 2 comments
#26 - a minor modification for download command
Issue -
State: closed - Opened by WeiChengTseng almost 4 years ago
- 1 comment
#25 - No module named 'dopamine.google'
Issue -
State: closed - Opened by jzitovsky about 4 years ago
- 2 comments
#24 - Why use a small batch size?
Issue -
State: closed - Opened by shengchao-y about 4 years ago
- 2 comments
#23 - How to evaluate agents?
Issue -
State: closed - Opened by Altriaex over 4 years ago
- 2 comments
#22 - Reading atari files directly.
Issue -
State: open - Opened by DuaneNielsen over 4 years ago
- 3 comments
Labels: documentation, enhancement
#21 - Python version?
Issue -
State: closed - Opened by johmicrot over 4 years ago
- 2 comments
#20 - The flag 'base_dir' is defined twice
Issue -
State: closed - Opened by Altriaex over 4 years ago
- 3 comments
#19 - Can a customized env be added to the current framework?
Issue -
State: closed - Opened by blurLake over 4 years ago
- 1 comment
#18 - Ambiguous selector
Issue -
State: closed - Opened by icomeaman over 4 years ago
- 1 comment
#17 - Getting 7 as action for a game with 3 actions
Issue -
State: closed - Opened by arjung128 over 4 years ago
- 1 comment
#16 - Code for offline continuous control experiments
Issue -
State: closed - Opened by CaralHsi over 4 years ago
- 1 comment
#15 - Data generation is very slow
Issue -
State: closed - Opened by muzerobot over 4 years ago
- 4 comments
#14 - Data Generation KeyError
Issue -
State: closed - Opened by muzerobot over 4 years ago
- 3 comments
#13 - Native Resolution Images
Issue -
State: closed - Opened by MatthewChang over 4 years ago
- 1 comment
#12 - Sequential Data
Issue -
State: closed - Opened by arjung128 over 4 years ago
- 2 comments
Labels: enhancement, good first issue
#11 - Raw results
Issue -
State: open - Opened by n17s over 4 years ago
- 8 comments
Labels: enhancement
#10 - How to train offline agent on the huge dataset (50 Million) ?
Issue -
State: open - Opened by LQNew almost 5 years ago
- 3 comments
Labels: documentation, good first issue, question
#9 - Can I train with my own dataset?
Issue -
State: closed - Opened by Marchen0 almost 5 years ago
- 4 comments
#8 - Retraining the online agent
Issue -
State: closed - Opened by n17s about 5 years ago
- 2 comments
#7 - Can i use the Tensorboard to check the intermediate results?
Issue -
State: closed - Opened by weihongwei0586 about 5 years ago
- 4 comments
#6 - Save the trained model to hdf5 file format
Issue -
State: closed - Opened by 2bben about 5 years ago
- 7 comments
#5 - TF Version must < 2.0?
Issue -
State: open - Opened by weihongwei0586 about 5 years ago
- 4 comments
Labels: good first issue
#4 - modify gs path and fix bug object has no attribute 'reorder_strategy'
Pull Request -
State: closed - Opened by tangbotony over 5 years ago
#3 - modify bucket path and fix MultiHeadDQNAgent bug
Pull Request -
State: closed - Opened by tangbotony over 5 years ago
- 1 comment
#2 - DQN Trained Model Releases
Issue -
State: closed - Opened by idigitopia over 5 years ago
- 2 comments
#1 - gsutil error
Issue -
State: closed - Opened by idigitopia over 5 years ago
- 2 comments