tqchen
Follow
Highlights
- Pro
Pinned
1,036 contributions in the last year
Contribution activity
March 2023
Created 86 commits in 6 repositories
Created 1 repository
- tqchen/web-stable-diffusion Python
Created a pull request in apache/tvm that received 1 comment
[Unity][WEBGPU] Codegen improvements and WebRuntime
This PR makes various improvements web codegen in relax web runtime. Correct support of shift operators. Update relax vm to make most use of intern…
+810
−171
•
1
comment
Opened 8 other pull requests in 2 repositories
apache/tvm
6
merged
mlc-ai/web-stable-diffusion
2
merged
Reviewed 47 pull requests in 5 repositories
apache/tvm
25 pull requests
- [Unity][Web] WebGPU explicit max buffer size
- [IR] Enhance IRModule SEqual/SHash to support cross function calls
- [Release] Adjust version to 0.11.1
- [TVMScript][Unity] Improve PyLint Compatibility
- [Unity][ci] Use CPU-SMALL instances
- [Transform] Use callable() instead of isinstance() for type checking
- [Unity][Transform] Let MetaScheduleApplyDatabase use DefaultSchedule when no record is found.
- [Unity] Allow user defined func attrs in emit_te
- [Unity][Op] Add repeat, tile, conv2d_transpose, avg_pool2d
-
[Unity][Op][Tweak] Improve
StructInfoinference forshape_of - Update to v0.12.dev0
- [Unity][WEB] Update text prompts for syntactical correctness
- [Unity][TVMScript] Enable Context-Aware Parsing
- [TIR] Enhance loop unroll with unroll local access
- [TVMScript] Sugar Var Definition in TIR Buffer
- [IR] Platform-independent SHash
- [Unity][Frontend] from_fx keeps parameters in order
- [Unity][Frontend] FX translator supports unwrapping unit return tuple
- [Unity][Frontend] Attach imported model weights, deprecate ImporterOutput
-
[Refactor] Move
VarUseDefAnalysisto header file - [Unity] Introduce Default GPU Schedule Pass
- [Unity][Frontend] FX translator support torch.baddbmm
- [Unity][TIR][Pass] ForceNarrowIndexToInt32
- [Unity][Fix] FX translating dtype
- [Fix][TIR] LowerCrossThreadReduction with write-back predicate
- Some pull request reviews not shown.
mlc-ai/web-stable-diffusion
7 pull requests
- Fix typo in stable_diffusion.js
- [Notebook] Merge walkthrough.ipynb and the TVMCon notebook
- [Notebook] TVMCon 2023 day 1 tutorial
- [Database] Update database with no spatial block loop unroll
- Walkthrough notebook and README with "get started"
- [Scheduler] Multistep DPM-solver for Python side
- [Web] Multistep DPM-solver for web side