Ph.D. student at Peking University. Interested in Reinforcement Learning & Multi-Agent Systems & Game Theory, and experienced in CV & NLP.
-
CFCS @ PKU
- Peking University, Beijing
Highlights
- Pro
- 1 discussion answered
Block or Report
Block or report XuehaiPan
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abuseHi there
Pinned
-
-
LaTeX-Templates Public
A collection of LaTeX templates in English/Chinese, with VS Code settings for LaTeX Workshop.
-
ray-project/ray Public
An open source framework that provides a simple, universal API for building distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyp…
-
1,823 contributions in the last year
Less
More
Contribution activity
April 2022
Created 16 commits in 3 repositories
Created a pull request in ray-project/ray that received 3 comments
[RLlib] Add missing methods on_trainer_init and on_sub_environment_created in MultiCallbacks
Why are these changes needed? Add the following missing methods to rllib.agent.callbacks.MultiCallbacks:
on_trainer_init (added in #22493)
on_sub…
+76
−41
•
3
comments
Opened 3 other pull requests in 3 repositories
ray-project/ray
1
open
XuehaiPan/nvitop
1
merged
Homebrew/install
1
open
Reviewed 5 pull requests in 3 repositories
openai/gym
2 pull requests
ray-project/ray
2 pull requests
Homebrew/install
1 pull request
Created an issue in ray-project/ray that received 1 comment
[Tune][Bug] tune.ExperimentAnalysis.best_checkpoint does not filter nan metric values
What happened + What you expected to happen
I'm using tune.run to train my RL policy and hope to get the best checkpoint weights for further analysis.
1
comment
Opened 3 other issues in 1 repository
157
contributions
in private repositories
Apr 1 – Apr 26