0
I Use This!
Moderate Activity

Commits : Listings

Analyzed 1 day ago. based on code collected 1 day ago.
Aug 16, 2024 — Aug 16, 2025
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Add torchao.dequantize_affine (#2558) More... 6 days ago
Add unbind_copy alias (#2557) More... 6 days ago
add (#2556) More... 6 days ago
Update Documentation Instructions (#2582) More... 10 days ago
Grammatical Doc Changes (#2567) More... 11 days ago
Fix for models (#2555) More... 16 days ago
Add linear as a supported op for activation quantization. (#2577) More... 16 days ago
Fix pytest-xdist compatibility in test_torch_ops (#2542) More... 18 days ago
Fix: Improve RangeDim default value logic and bound validation [#2560] (#2561) More... 18 days ago
Beta install instructions (#2562) More... 20 days ago
9.0b1 Release (#2559) More... 20 days ago
Get rid of dependency on distutils (#2540) More... 2 months ago
Fix typos (#2528) More... 2 months ago
TF 1.x warning (#2531) More... 2 months ago
Add warning about droping Python 3.7 support (#2529) More... 2 months ago
This PR addresses two key issues. First, it resolves a datatype inconsistency by explicitly specifying the dtype when generating input data during benchmarking, ensuring reliable behavior when inputs are not provided. Second, it fixes a problem in the MLModelAsyncWrapper where the source model directory was inadvertently deleted when initializing the wrapper with a model path instead of a specification. (#2534) More... 2 months ago
Fix complex const initialization (fixes #2530) (#2532) More... 2 months ago
Fix failing MLModelValidator tests (#2533) More... 2 months ago
fix: ensure x/y inputs are non-const before adding q/dq pair (#2526) More... 3 months ago
Optimize torch im2col implementation (#2523) More... 3 months ago
Adds conversion date to model meta data. (#2521) More... 3 months ago
docs: update cto.coreml path in ptq docs (#2513) More... 3 months ago
Fix import in Debugging And Perf Utilities guide (#2508) More... 3 months ago
Add experimental tag (#2505) More... 3 months ago
fix: numpy 2 prefers asarray() instead of copy=False (#2488) More... 3 months ago
Clean Ups (#2499) More... 3 months ago
Adds allowLowPrecisionAccumulationOnGPU optimization hint to MLModel and CompiledMLModel (#2498) More... 4 months ago
Add documentation for debugging and performance utilities (#2496) More... 4 months ago
Fix the reflection padding optimization when slice_by_index's `end_mask` is not specified (#2489) More... 4 months ago
8.3 release (#2492) More... 4 months ago