strisunshinewentingwang/pytorch

Pull Requests Commits

[quant] Finishing refactor for quantization test files (#37366)

jerryzh168 committed 4 years ago

facdd15c

[quant][graph] Add JIT passes for dynamic quant multi uses of quant node (#37125)

supriyar committed 4 years ago

e69115ec

[quant][graph] Run dynamic quantization for specific ops (#37093)

supriyar committed 4 years ago

3b9ddab0

[graph][quant] Enable accessing child/grandchild modules in forward (#37045)

supriyar committed 4 years ago

9dab3ed5

[quant][graph] Add check for qconfig_dict key (#37014)

supriyar committed 4 years ago

e55d2e6f

[jit] Fix pretty printing of functions (#37432)

suo committed 4 years ago

92b9089f

Move DistributonTemplates to anonymous namespace (#37429)

malfet committed 4 years ago

07bb442b

Don't use NonVariableTypeMode in custom ops (#37355)

smessmer committed 4 years ago

12f5a328

run the simple executor for jit tests by default, add profiling jobs … (#37017)

Krovatkin committed 4 years ago

edc5ef1a

[jit] __deepcopy__ for `RecursiveScriptModule` (#32684)

jerryzh168 committed 4 years ago

6fa76b8a

Retry anaconda upload (#37414)

peterjc123 committed 4 years ago

e5a24a63

Fix `TensorIterator::view_offsets_` size (#37214)

Emilio Castillo committed 4 years ago

273c4641

Revert D21286660: [quant] Generalizing _calculate_dynamic_qparams in quantized test

Mike Ruberry committed 4 years ago

dcd8a1b3

Remove ONNX BatchNorm(12) test and converter. (#37309)

codemzs committed 4 years ago

6c0f447b

[pytorch] add '__BASE__' section to op deps to factor out frequently used util ops (#37404)

ljk53 committed 4 years ago

8258d42b

[pytorch] remove unused flags from code analyzer & move format support to python (#37393)

ljk53 committed 4 years ago

e0a5b443

[quant] Generalizing _calculate_dynamic_qparams in quantized test (#37451)

z-a-f committed 4 years ago

239ce75a

Resubmit "Fix NaN error in dynamic quantization in qLinear, re-enable test_quantized_rnn"" (#37458)

raghuramank100 committed 4 years ago

024f663f

[caffe2] Copy tensor in single tensor input case in UnPackRecordsOp (#37454)

Hao Lu committed 4 years ago

5b6f6da1

Remove Python 2 string compatibility in ATen/function_wrapper.py (#37388)

xuhdev committed 4 years ago

d1a39815

[ONNX] Adding 'numel' and 'to' export for script module (#36501)

neginraoof committed 4 years ago

58092880

Synchronize MAGMA functions with the current CUDA stream (#36605)

peterbell10 committed 4 years ago

a51f047c

[resubmit] Enable global observers API (#37382)

Ilia Cherniavskii committed 4 years ago

d068a456

[hotfix] Workaround for older versions of ninja (#37417)

bwasti committed 4 years ago

4234d624

Replacing EHa with EHsc (#37235)

peterjc123 committed 4 years ago

c5d6f59a

Fixes type annotations for named tensors #27846 (#36890)

physicsrob committed 4 years ago

8fe2a5e9

[Bazel] Build `ATen_CPU_AVX2` lib with AVX2 arch flags enabled (#37381)

malfet committed 4 years ago

ebcacd5e

remove record_function_enter and record_function_exit from header (#37052)

rohan-varma committed 4 years ago

b37080d9

[caffe2] Fast path for single tensor in UnPackRecordsOp (#37361)

Hao Lu committed 4 years ago

48b126f4

Reduce volume of spammy warning (#37360)

Ilia Cherniavskii committed 4 years ago

da64ed14

Newer Older