..
custom
Better msg ( #43848 )
2020-08-31 10:36:59 -07:00
nql
Replace c10::guts::stuff with std::stuff ( #30915 )
2019-12-16 13:57:19 -08:00
annotations.cc
fix warnings reported by PVS ( #33868 )
2020-03-02 18:51:38 -08:00
annotations.h
backend_cutting_test.cc
Create BackendTransformerBase to host common functions used for backend lowering ( #17074 )
2019-02-14 17:57:03 -08:00
backend_cutting.cc
Fix typos, via a Levenshtein-type corrector ( #31523 )
2020-01-17 16:03:19 -08:00
backend_cutting.h
Fuse to individual operators to GatherFuse8BitRowwiseQuantFloatMulLengthElim ( #25519 )
2019-09-03 19:08:49 -07:00
backend_transformer_base.cc
Override shape hints with real weight shape extracted from workspace ( #40872 )
2020-07-02 12:55:29 -07:00
backend_transformer_base.h
Changed Blacklisted to Blocklisted ( #42100 )
2020-07-28 13:21:26 -07:00
bound_shape_inference_test.cc
Add shape inference to SparseLengthsSumSparse ops ( #43181 )
2020-08-18 09:36:53 -07:00
bound_shape_inferencer.cc
Add operator to compute the equalization scale ( #45096 )
2020-09-24 15:19:49 -07:00
bound_shape_inferencer.h
[Onnxifi] Support running with quantized int8 inputs ( #41820 )
2020-07-22 13:42:34 -07:00
CMakeLists.txt
converter_nomigraph_test.cc
[nomnigraph] Copy device option when customize the op conversion ( #34976 )
2020-03-19 22:48:28 -07:00
converter.cc
[nomnigraph] Copy device option when customize the op conversion ( #34976 )
2020-03-19 22:48:28 -07:00
converter.h
[nomnigraph] Copy device option when customize the op conversion ( #34976 )
2020-03-19 22:48:28 -07:00
dead_code_elim_test.cc
dead_code_elim.cc
device_test.cc
device.cc
device.h
distributed_converter.cc
distributed_test.cc
distributed.cc
distributed.h
fakefp16_transform.cc
fuse layernorm + quantize ( #44232 )
2020-09-12 13:32:33 -07:00
fakefp16_transform.h
add fake fp16 fusions to net transforms ( #42927 )
2020-08-14 13:30:27 -07:00
fusion.cc
fusion.h
glow_net_transform.cc
add fake fp16 fusions to net transforms ( #42927 )
2020-08-14 13:30:27 -07:00
glow_net_transform.h
add fake fp16 fusions to net transforms ( #42927 )
2020-08-14 13:30:27 -07:00
mobile_test.cc
mobile.cc
caffe2 | Revert range loop analysis fix
2020-06-27 13:02:23 -07:00
mobile.h
onnx_convert.h
onnxifi_op.cc
Add API for onnxifi with AOT Glow ONNX ( #44021 )
2020-09-03 22:46:20 -07:00
onnxifi_op.h
Add GlowLoadAOTModel flag ( #45189 )
2020-09-23 20:50:04 -07:00
onnxifi_transformer.cc
fuse layernorm + quantize ( #44232 )
2020-09-12 13:32:33 -07:00
onnxifi_transformer.h
Blacklist to Blocklist in onnxifi_transformer ( #42590 )
2020-08-07 08:05:32 -07:00
optimize_ideep.cc
Upgrade MKL-DNN to DNNL v1.2 ( #32422 )
2020-03-26 22:07:59 -07:00
optimize_ideep.h
support pre-convert filter format for mkldnn training mode and change 'OptimizeForIdeep' to 'OptimizeForMkldnn' ( #15171 )
2019-03-29 19:00:48 -07:00
optimizer.cc
optimizer.h
passes.cc
passes.h
shape_info.cc
[caffe2] Support data types in shape hints ( #45110 )
2020-09-22 17:49:33 -07:00
shape_info.h
Expose a generic shape info struct for ONNXIFI Python interface ( #42421 )
2020-08-03 16:10:05 -07:00
split_slss_test.cc
Add shape inference to SparseLengthsSumSparse ops ( #43181 )
2020-08-18 09:36:53 -07:00
tvm_transformer.cc
Fixed for caffe2/opt/tvm_transformer.cc ( #44249 )
2020-09-18 00:03:59 -07:00
tvm_transformer.h
Fixed for caffe2/opt/tvm_transformer.cc ( #44249 )
2020-09-18 00:03:59 -07:00
1539d4a664