YuliangLiu0306

Follow

YuliangLiu0306

Follow

19 followers · 3 following

Achievements

BetaSend feedback

Achievements

BetaSend feedback

Block or Report

Block or report YuliangLiu0306

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Pinned

ColossalAI Public

Forked from hpcaitech/ColossalAI

Colossal-AI: A Unified Deep Learning System for Large-Scale Parallel Training

Python 1

747 contributions in the last year

Learn how we count contributions

Activity overview

Contributed to hpcaitech/ColossalAI, hpcaitech/ColossalAI-Examples, hpcaitech/Titans and 7 other repositories

Contribution activity

March 2023

Created 17 commits in 1 repository

hpcaitech/ColossalAI 17 commits

Created a pull request in hpcaitech/ColossalAI that received 5 comments

[DTensor] refactor dtensor with new components

📌 Checklist before creating the PR I have created an issue for this PR for traceability The title follows the standard format: [doc/gemini/tenso…

+20 −41 • 5 comments

Opened 15 other pull requests in 1 repository

hpcaitech/ColossalAI 14 merged 1 closed

[hotfix] meta_tensor_compatibility_with_torch2 Mar 30
[examples] polish AutoParallel readme Mar 28
[autoparallel] adapt autoparallel with new analyzer Mar 27
[API] implement device mesh manager Mar 23
[hotfix] skip torchaudio tracing test Mar 23
[Analyzer] fix analyzer tests Mar 22
[hotfix] layout converting issue Mar 21
[FX] refactor experimental tracer and adapt it with hf models Mar 17
[test] add torchrec models to test model zoo Mar 15
[API] implement environment table Mar 10
[DTensor] implement layout converter Mar 8
[DTensor] refactor CommSpec Mar 7
[hotfix] skip auto checkpointing tests Mar 7
[DTensor] refactor sharding spec Mar 3
[hotfix] add shard dim to aviod backward communication error Mar 1

Reviewed 16 pull requests in 1 repository

hpcaitech/ColossalAI 16 pull requests

[booster] implement Gemini plugin Mar 31
[WIP] fix torch 2.0 compatibility Mar 30
[hotfix] fix torch 2.0 compatibility in installation denpendencies Mar 30
[fx] meta registration compatibility Mar 27
[booster] implemented the torch ddd + resnet example Mar 27
[hotfix] polish code style of stable diffusion example Mar 24
[lazyinit] combine lazy tensor with dtensor Mar 22
[booster] implemented the cluster module Mar 22
[booster] added the plugin base and torch ddp plugin Mar 21
[FX] refactor experimental tracer and adapt it with hf models Mar 21
[auto-parallel] add auto-offload feature Mar 21
[test] fixed torchrec registration in model zoo Mar 20
[pipeline] Add Simplified Alpa DP Partition Mar 7
[analyzer] a minimal implementation of static graph analyzer Mar 7
[autoparallel] stable and user-friendly of auto-ckpt solvers Mar 7
[DTensor] implementation of dtensor Mar 1

Opened 8 issues in 1 repository

hpcaitech/ColossalAI 2 open 6 closed

[FEATURE]: Integrate the latest analyzer into the automatic parallelism module Mar 28
[FEATURE]: fix analyzer tests for torch 1.12.1 Mar 22
[FEATURE]: integrate tracer Mar 17
[FEATURE]: Refactor DTensor Mar 10
[FEATURE]: Implement LayoutConverter Mar 9
[FEATURE]: Refactor CommSpec Mar 7
[FEATURE]: Refactor ShardingSpec Mar 3
[FEATURE]: dtensor demo Mar 1

Started 1 discussion in 1 repository

hpcaitech/ColossalAI

DTensor implementation
This contribution was made on Mar 13