tvm
https://github.com/apache/tvm
Python
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported3 Subscribers
Add a CodeTriage badge to tvm
Help out
- Issues
- [Bug][Relax] `FoldConstant` leaks bindings emitted by a legalizer it only tries speculatively: `dynamic_strided_slice` leaves a `call_tir` to an undefined `shape_func` ("There is no definition of Module.shape_func")
- [Bug] layer_norm should support broadcast in all axes
- [DLight] Schedule shallow reductions serially per thread
- [Bug][TIRx][LLVM] Non-unit-stride masked load emits an invalid LLVM intrinsic
- [Bug][TIRx][TensorFlattener] Flattened masked load retains rank-two indices
- [Relax][TFLite] Flatten FC input without keep_num_dims
- [Fix][Relax] Normalize squeeze axes before layout inference
- [Fix][DLight] Decline matmul schedules with unsupported consumer chains
- [Relax][ONNX] Add Swish and Celu operator converters
- [Relax][Frontend] Add missing frontend operators: ONNX Swish/ReverseSequence and TFlite DYNAMIC_UPDATE_SLICE
- Docs
- Python not yet supported