opendilab/DI-engine issues and pull requests

#717 - feature(rjy): add HAPPO algorithm

Pull Request - State: closed - Opened by nighood about 1 year ago - 1 comment
Labels: algo

#716 - why no tanh after sample action in ppo

Issue - State: closed - Opened by zhixiongzh about 1 year ago - 4 comments
Labels: discussion, algo

#715 - how to use logger

Issue - State: closed - Opened by zhixiongzh about 1 year ago - 2 comments
Labels: good first issue

#714 - not compatible with gym_pybullet_drones env

Issue - State: closed - Opened by zhixiongzh about 1 year ago - 5 comments
Labels: bug, env

#713 - fix(luyd): fix new pipeline impala in Lunarlander and Atari env

Pull Request - State: closed - Opened by AltmanD about 1 year ago - 2 comments
Labels: bug, algo

#712 - feature(lxy): add dropout layers to dqn

Pull Request - State: closed - Opened by karroyan about 1 year ago
Labels: enhancement

#711 - fix(zjow): fix mujoco-py compatibility issue for cython<3

Pull Request - State: closed - Opened by zjowowen about 1 year ago
Labels: bug, docker

#710 - NameNotFound: Environment stocks doesn't exist.

Issue - State: closed - Opened by prowgrammmer about 1 year ago - 11 comments
Labels: bug

#709 - Error occurs when running in parallel

Issue - State: closed - Opened by weixians about 1 year ago - 2 comments
Labels: bug, parallel-dist

#708 - fix(lixuelin): to_ndarray fails to assign dtype for scalars

Pull Request - State: closed - Opened by lxl2gf about 1 year ago
Labels: bug

#707 - Error occurs when async_mode=True in example/ppo.py

Issue - State: closed - Opened by weixians about 1 year ago - 3 comments
Labels: bug

#706 - feature(whl): add example of dqn eval

Pull Request - State: closed - Opened by kxzxvbk about 1 year ago - 1 comment
Labels: enhancement

#704 - fix(zyz): fix type spell error

Pull Request - State: closed - Opened by Privilger about 1 year ago - 2 comments

#703 - feature(pu): add three variants of Bilinear classes and a FiLM class

Pull Request - State: closed - Opened by puyuan1996 about 1 year ago
Labels: enhancement, algo

#702 - polish(nyz): polish atari envpool demo

Pull Request - State: closed - Opened by PaParaZz1 about 1 year ago - 2 comments
Labels: env

#701 - feature(cxy): add iql policy

Pull Request - State: closed - Opened by Mossforest about 1 year ago
Labels: algo

#700 - feature(zc): add plan diffuser

Pull Request - State: closed - Opened by Super1ce about 1 year ago - 1 comment
Labels: algo

#699 - test

Pull Request - State: closed - Opened by wey-code about 1 year ago

#697 - Halfcheetah has negative state in z position

Issue - State: closed - Opened by zhixiongzh about 1 year ago - 2 comments
Labels: env

#696 - feature(zjow): add middleware for ape-x structure pipeline

Pull Request - State: closed - Opened by zjowowen about 1 year ago - 1 comment
Labels: efficiency optimization, enhancement

#695 - doc(zjow): update README.md with openxlab badge

Pull Request - State: closed - Opened by zjowowen about 1 year ago
Labels: doc

#694 - Setting of Parameters in gym_anytrading and How to Evaluate Training Performance

Issue - State: closed - Opened by 123xian123 about 1 year ago - 3 comments
Labels: config

#693 - refactor(lyd): refactor dt_policy in new pipeline and add img input support

Pull Request - State: closed - Opened by AltmanD about 1 year ago
Labels: refactor

#692 - `GAIL` the algorithm performs much worse than from its original paper.

Issue - State: closed - Opened by shenpengfii about 1 year ago - 6 comments
Labels: algo, config

#691 - evaluate on same env for multiple time after fails

Issue - State: closed - Opened by zhixiongzh about 1 year ago - 4 comments
Labels: discussion

#690 - feature(nyz): add new pipeline impala demo

Pull Request - State: closed - Opened by PaParaZz1 about 1 year ago - 1 comment
Labels: algo

#689 - Does model support multiple input parameters?

Issue - State: closed - Opened by webbery about 1 year ago - 1 comment
Labels: good first issue, discussion, algo

#688 - `enable_save_figure()` has no detailed document, and its comment is wrong.

Issue - State: closed - Opened by shenpengfii about 1 year ago - 4 comments
Labels: discussion, env

#687 - refactor(lyd): refactor dt_policy in new pipeline

Pull Request - State: closed - Opened by AltmanD about 1 year ago
Labels: algo, refactor

#686 - Env manager is closed after wandb logger is set

Issue - State: closed - Opened by webbery about 1 year ago - 2 comments
Labels: bug

#685 - wandb crash

Issue - State: closed - Opened by webbery about 1 year ago - 1 comment
Labels: bug

#684 - feature(rjy): add ptz-mpe-simple_tag env

Pull Request - State: closed - Opened by nighood about 1 year ago
Labels: env

#683 - feature(cxy): add averaged-dqn policy

Pull Request - State: open - Opened by Mossforest about 1 year ago
Labels: algo

#682 - Document about config

Issue - State: closed - Opened by webbery about 1 year ago - 6 comments
Labels: good first issue, config

#681 - Question about MBPO.

Issue - State: closed - Opened by Uptomylimit over 1 year ago - 5 comments
Labels: bug, config

#680 - feature(yzj): add scatter connection model

Pull Request - State: closed - Opened by jayyoung0802 over 1 year ago
Labels: algo

#679 - polish(nyz): polish offpolicy RL multi-gpu DDP training

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago - 1 comment
Labels: efficiency optimization, enhancement

#678 - feature(nyz): fix py37 macos ci bug and update default pytorch to 1.12.1

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago
Labels: docker

#677 - feature(cxy): add cliffwalking env

Pull Request - State: closed - Opened by Mossforest over 1 year ago
Labels: env

#676 - `dqn_rnd.py` example appears to be incorrect

Issue - State: closed - Opened by ahalev over 1 year ago - 2 comments
Labels: bug, discussion

#675 - feature(whl): add SIL policy

Pull Request - State: open - Opened by kxzxvbk over 1 year ago - 1 comment
Labels: algo

#674 - Doc(zjow): Huggingface badge

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: doc

#673 - feature(cy): add tensor stream merging tools

Pull Request - State: closed - Opened by Cloud-Pku over 1 year ago - 1 comment
Labels: enhancement

#672 - polish(nyz): simplify requirements

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago - 1 comment
Labels: enhancement, test

#671 - WIP: fix(pu): fix decision_transformer encoder_hidden_size_list bug, polish dt configs

Pull Request - State: closed - Opened by puyuan1996 over 1 year ago
Labels: algo, config

#670 - How to run the example of Decision Transformer

Issue - State: closed - Opened by shimoshida over 1 year ago - 1 comment
Labels: bug, good first issue, config

#669 - Implementation of model-based RL algorithm Dreamer

Issue - State: closed - Opened by openRiemann over 1 year ago - 3 comments
Labels: discussion, algo

#668 - failed to run gym_hybrid_pdqn_config.py

Issue - State: closed - Opened by super1603 over 1 year ago - 2 comments
Labels: duplicate

#667 - feature(whl): add tabmwp env and prompt pg policy

Pull Request - State: closed - Opened by kxzxvbk over 1 year ago
Labels: algo, data

#666 - the codebase design has high coupling and need to be refactored

Issue - State: closed - Opened by walkacross over 1 year ago - 3 comments
Labels: discussion

#665 - The distributed demo for league training

Issue - State: closed - Opened by hanshuo123i over 1 year ago
Labels: enhancement, parallel-dist

#664 - HELP: failed to run gym_hybrid_pdqn_config..py

Issue - State: closed - Opened by xujinming01 over 1 year ago - 9 comments
Labels: bug

#662 - fix(zp): add record_random_collect option

Pull Request - State: closed - Opened by zhangpaipai over 1 year ago - 1 comment
Labels: enhancement

#661 - feature(nyz): add MAPPO/MASAC task example

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago - 1 comment
Labels: algo, config

#660 - polish(pu): add LN and GN norm_type support in ResBlock

Pull Request - State: closed - Opened by puyuan1996 over 1 year ago
Labels: enhancement

#659 - fix(zjow): update td3bc d4rl config

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: config

#658 - fix(lxy): add normal value norm baseline

Pull Request - State: closed - Opened by karroyan over 1 year ago
Labels: algo

#657 - 使用自己的专家轨迹

Issue - State: closed - Opened by nianba over 1 year ago - 2 comments
Labels: good first issue, data

#656 - feature(zjow): add torch_save method in case of disk being full

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: enhancement

#655 - fix(zjow): fix atari env id

Pull Request - State: closed - Opened by zjowowen over 1 year ago
Labels: config

#654 - fix(eltociear): typo in gyn any_trading env

Pull Request - State: closed - Opened by eltociear over 1 year ago

#653 - fix(zjow): fix incompatible gym version bug in Dockerfile.env

Pull Request - State: closed - Opened by zjowowen over 1 year ago
Labels: bug

#652 - feature(zp): add dreamerv3 algorithm

Pull Request - State: closed - Opened by zhangpaipai over 1 year ago - 1 comment
Labels: algo

#651 - feature(wgt): add profiling stuff

Pull Request - State: closed - Opened by SolenoidWGT over 1 year ago
Labels: efficiency optimization

#650 - fix(pu): fix last_linear_layer_weight_bias_init_zero in MLP and add its unittest

Pull Request - State: closed - Opened by puyuan1996 over 1 year ago - 1 comment
Labels: bug, test

#649 - average value in log

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 2 comments
Labels: good first issue, discussion

#648 - how to add custom loss in tensorboard

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 1 comment

#647 - fix(zjow): fix train_data type bug in trainer middleware

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: bug, P0

#646 - 运行DI-Zoo里面写好的config报错

Issue - State: closed - Opened by xujinming01 over 1 year ago - 2 comments
Labels: bug, config

#645 - Unresolved for 18 Months: Issue #90 - League Training for SlimeVolleyball Env

Issue - State: closed - Opened by zxzzz0 over 1 year ago - 4 comments
Labels: env

#644 - feature(zjow): add PPO demo for complex env observation

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: algo

#643 - Placing restrictions on the agent for trading

Issue - State: closed - Opened by codecolony over 1 year ago - 1 comment
Labels: env

#642 - feature(whl): Oneflow Pipeline for DQN+cartpole

Pull Request - State: closed - Opened by kxzxvbk over 1 year ago - 1 comment
Labels: enhancement

#641 - fix(nyz): fix confusing shallow copy operation bug about next_obs

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago - 1 comment
Labels: bug

#640 - feature(zc): add bcq algorithm

Pull Request - State: closed - Opened by Super1ce over 1 year ago - 2 comments
Labels: algo

#639 - feature(zc): add EDAC and modify config of td3bc

Pull Request - State: closed - Opened by Super1ce over 1 year ago
Labels: algo

#638 - redundant operation in to_tensor_transitions [bug]

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 2 comments
Labels: bug, discussion

#637 - feature(zjow): add new pipeline agent sac/ddpg/a2c

Pull Request - State: closed - Opened by zjowowen over 1 year ago - 1 comment
Labels: enhancement, algo

#636 - refactor(gry): refactor reward model

Pull Request - State: open - Opened by ruoyuGao over 1 year ago - 1 comment
Labels: refactor

#635 - discrepancy between the expected(doc) and actual results with Pong+DQN

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 8 comments
Labels: reproducibility

#634 - can't use async_mode for with task.start(async_mode=True, ctx=OnlineRLContext())

Issue - State: closed - Opened by laijiajian over 1 year ago - 4 comments
Labels: parallel-dist

#633 - Unify main_config and create_config into one config

Issue - State: closed - Opened by way-zer over 1 year ago - 1 comment
Labels: discussion, config

#632 - the use of BaseSerialCommander and ppo_command

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 3 comments
Labels: good first issue, discussion

#631 - fix(pu): fix test_simple_spread_v2, add ptz eval config, add check_ding_env_wrapper

Pull Request - State: closed - Opened by puyuan1996 over 1 year ago - 1 comment
Labels: env

#630 - PPOOffpolicy _forward_learn method uses only 1-step value instead of n-step value

Issue - State: closed - Opened by luxiaolei over 1 year ago - 2 comments
Labels: discussion

#629 - 求助！

Issue - State: closed - Opened by sdsafsdfsaf over 1 year ago - 1 comment
Labels: bug

#628 - feature(wgt): add torch-rpc fix dockerfile

Pull Request - State: closed - Opened by SolenoidWGT over 1 year ago - 1 comment
Labels: docker

#627 - Missing of Regularization of GCL

Issue - State: closed - Opened by zhixiongzh over 1 year ago - 1 comment
Labels: discussion

#626 - feature(zc): add EDAC and modify config of td3bc

Pull Request - State: closed - Opened by Super1ce over 1 year ago - 1 comment
Labels: algo

#625 - feature(whl): add gpt utils

Pull Request - State: closed - Opened by kxzxvbk over 1 year ago - 1 comment
Labels: enhancement, docker

#624 - polish(gry): polish reward model and td error

Pull Request - State: closed - Opened by ruoyuGao over 1 year ago - 1 comment
Labels: enhancement

#623 - subprocess_env_manager has an error

Issue - State: closed - Opened by youhu868 over 1 year ago - 2 comments
Labels: bug

#622 - 安装过程中，执行pip install DI-engine[common_env,test,fast] 报错

Issue - State: closed - Opened by sdsafsdfsaf over 1 year ago

#621 - Call for contributors!

Issue - State: open - Opened by PaParaZz1 over 1 year ago - 9 comments
Labels: good first issue, discussion

#620 - refactor(nyz): remove policy cfg sub fields requirements

Pull Request - State: closed - Opened by PaParaZz1 over 1 year ago
Labels: refactor

#619 - feature(zp): add bonus/c51

Pull Request - State: closed - Opened by zhangpaipai over 1 year ago - 1 comment
Labels: enhancement

#618 - Does TransformerMemoryWrapper have a logical error in resetting memories

Issue - State: closed - Opened by youhu868 over 1 year ago - 2 comments
Labels: discussion

#617 - fix(jimmy): fix typo in task

Pull Request - State: closed - Opened by jimmydengpeng over 1 year ago - 1 comment
Labels: typo

#616 - The learn model of PPO policy has a reset error

Issue - State: closed - Opened by youhu868 over 1 year ago - 4 comments
Labels: bug

#614 - Installing using pip results in package resolution impossible error for pettingzoo and torch

Issue - State: closed - Opened by almair404 over 1 year ago - 2 comments
Labels: bug

GitHub / opendilab/DI-engine issues and pull requests