Focusing
Highlights
- Pro
Block or Report
Block or report tqchen
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
apache/tvm Public
Open deep learning compiler stack for cpu, gpu and specialized accelerators
-
dmlc/xgboost Public
Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library, for Python, R, Java, Scala, C++ and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow
-
apache/mxnet Public
Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
1,222 contributions in the last year
Less
More
Contribution activity
April 2023
Created 66 commits in 5 repositories
Created 3 repositories
- tqchen/web-llm Python
- tqchen/sentencepiece-js TypeScript
- tqchen/tokenizers-wasm Rust
Created a pull request in apache/tvm that received 13 comments
[ARITH][BUGFIX] Fix a bug of iter map floormod(x,2) simplify
This PR fixes a previous bug introduced in itermap detection. Specifically, y - (x % 2) were simplified to y + (x % 2) - 1. Which is wrong. The wor…
+86
−65
•
13
comments
Opened 16 other pull requests in 2 repositories
apache/tvm
15
merged
- [Unity][TARGET] Updates vulkan codegen for DeclBuffer
- [Unity] Improve error message in webgpu request
- [Unity] hotfix webgpu codegen for vec load
- [CODEGEN] Fix metal codegen when with only single working dim
- [Unity][CODEGEN] Fix metal codegen when with only single working dim
- [Unity] Update specific builtins for LM
- [Unity] Fix ForceNarrowI32 with pod arguments
- [ARITH][TensorIR] Improve CompactBufferRegion for symbolic shape
- [TensorIR] Fix ComputeAt with perfect symbolic bound
- [Unity] Enable pod args in WebGPU
- [Unity] Fix emit_te with symbolic input
- [ARITH] Enhance IterMapSimplify for symbolic
- [ARITH] Enhance CanonicalSimplify to Simplify ProdDiv
- [ARITH] Enhance CanProve to handle symbolic bound
- [Unity][VM] Add Attention KV cache builtin
mlc-ai/web-llm
1
merged
Reviewed 58 pull requests in 5 repositories
apache/tvm
25 pull requests
-
[Unity]
enable_warningoption for LegalizeOps and MSApplyDatabase - Adding powerPreference argument to navigator.gpu.requestAdapter
- Fix bug about wrong attribute name
- [Unity] update ci cpu/gpu images
- [Unity] BlockBuilder assigning unique tensor names in call_te
- [Unity][Frontend] Some changes on the PyTorch FX Frontend
- [Unity][VM] LibComparator using dtype from input
-
[Unity][Bugfix] Resolve failure on
test_e2e_op_dynamic.py - [Unity][VM] Converting tuple arg to Python tuple
- [Node] Allow alternative root names in ObjectPath::Root()
- [CI] Add JAX deps in Dockerfiles
- [Unity][PyTorch] Disable gradient during dynamo subgraph capture to save RAM
- [ARITH][TensorIR] Improve CompactBufferRegion for symbolic shape
- [Fix][TIR][Analysis] Reduction block checking alloc_buffers
- [TIR][Schedule] Method returning the function being worked on
- [Docs] Fix typo in the Vitis AI Integration docs
- [Unity][IR] Purity Tracking
- [ARITH][BUGFIX] Fix a bug of iter map floormod(x,2) simplify
- [Unity][MetaSchedule] Skip Scheduled PrimFuncs in Task Generation
- [Fix][TIR] UnifyThreadBinding creating unit loop with annotation
- [Unity][TVMScript] Update struct_info for GlobalVar
- [Unity] Enhance Dynamic-aware FuseTIR
- [Arith] Simplifications for floormod(x, 2)
- [Unity][Op] add einsum and flip in Relax
- [Unity][Op] Symbolic shape support of take grad
- Some pull request reviews not shown.






