Skip to content

Pull requests: opendilab/DI-engine

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Sort

Pull requests list

feature(zjow): add Implicit Q-Learning algo Add new algorithm or improve old one
#821 opened Jul 29, 2024 by zjowowen Loading…
feature(wrh): add EDT code algo Add new algorithm or improve old one
#808 opened Jun 20, 2024 by ruiheng123 Loading…
3 tasks
feature(xrk): add q-transformer algo Add new algorithm or improve old one
#783 opened Mar 22, 2024 by rongkunxue Loading…
3 tasks
feature(zc): add MetaDiffuser and prompt-dt algo Add new algorithm or improve old one
#771 opened Jan 30, 2024 by Super1ce Loading…
feature(zjow): add envpool new pipeline enhancement New feature or request
#753 opened Nov 24, 2023 by zjowowen Loading…
feature(whl): add rlhf pipeline. algo Add new algorithm or improve old one enhancement New feature or request
#748 opened Nov 6, 2023 by kxzxvbk Loading…
3 tasks
feature(cxy): add averaged-dqn policy algo Add new algorithm or improve old one
#683 opened Jul 8, 2023 by Mossforest Loading…
5 tasks
feature(whl): add SIL policy algo Add new algorithm or improve old one
#675 opened Jun 9, 2023 by kxzxvbk Loading…
3 tasks
refactor(gry): refactor reward model refactor refactor module or component
#636 opened Apr 5, 2023 by ruoyuGao Loading…
1 of 3 tasks
feature(whl): add PC+MCTS code algo Add new algorithm or improve old one
#603 opened Mar 5, 2023 by kxzxvbk Loading…
3 tasks
feature(wgt): enable DI using torch-rpc to support GPU-p2p and RDMA-rpc efficiency optimization Efficiency optimization (time, memory and so on)
#562 opened Dec 25, 2022 by SolenoidWGT Loading…
2 of 3 tasks
feature(zms): add new league middlewares and other models and tools. enhancement New feature or request
#458 opened Aug 26, 2022 by hiha3456 Loading…
3 tasks
ProTip! Follow long discussions with comments:>50.