zjgarvey 694210f429 [TorchToLinalg] Fix Quantized Convolution Accumulator Type (#3459) ... 1. truncates zero-points to i32 2. modifies the default accumulator type for i8 from i64 to i32. 3. now uses the input dtype to infer accumulator dtype.		2024-06-20 13:54:20 -07:00
..
CMakeLists.txt	[sparse] propagate sparsity properly when decompose torch operations. (#3318)	2024-05-15 10:09:27 -07:00
SparsityUtils.cpp	[NFC] Remove unused header files (#3386)	2024-05-30 14:30:36 +08:00
TorchUpstream.cpp	[ONNX] Add OnnxToTorch lowering for Onnx.NegativeLogLikelihoodLoss Op (#3380)	2024-06-14 22:01:11 +05:30
Utils.cpp	[TorchToLinalg] Fix Quantized Convolution Accumulator Type (#3459)	2024-06-20 13:54:20 -07:00