microsoft/onnxruntime

Pull Requests Commits

Remove unused variable. (#11043)

edgchen1 committed 4 years ago

Verified 8f456735

Enabled Cast operator on OneDNN EP (#11023)

Erick Alejandro Muñoz Alvarado committed 4 years ago

Verified 6c005bfd

Fuse LayerNormalization for Apex O2 (#10233)

Lafi7e committed 4 years ago

Verified 6a6840d5

[CUDA] Optimize Conv and ConvGrad for Training (#10999)

Lafi7e committed 4 years ago

Verified 3b6cee80

Bump master version to 1.12 (#10797)

chilo-ms committed 4 years ago

Verified 8ba52b0a

Move node EP assignment for ORT format into SessionState::FinalizeSessionState() (#10944)

edgchen1 committed 4 years ago

Verified 93714017

Add utility to get the gradient graph from GradientGraphBuilder (#10995)

baijumeswani committed 4 years ago

Verified 9c6cc018

Symmetric Quant indirect Conv kernel for ARMv8 A55 chip (#10862)

chenfucn committed 4 years ago

Verified dc721591

Add linux and macos arm64 java aritifacts (#10981)

leqiao-1 committed 4 years ago

Verified 8ddc45f5

[cmake] Add keyword STATIC to add_library in function onnxruntime_add_static_library (#10998)

JackBoosY committed 4 years ago

Verified d1be71ea

Fixed creation of ORT_Value to pass offset of 0 (#11004)

chandru-r committed 4 years ago

Verified cb31b7ea

Clarify usage of kOnnxDomainAlias. (#10962)

skottmckay committed 4 years ago

Verified 47c09e67

Improve NonZero on CUDA/ROCM (#10307)

pengwa committed 4 years ago

Verified 89ef987a

Adding support for saving and loading train step info properties in the state dict and checkpoint file. (#10569)

mpapdiwala committed 4 years ago

Verified 1e917c87

Update docstrings in quantize.py (#10952)

chausner-audeering committed 4 years ago

Verified 989e6400

register custom_op_symbolic for squeeze (#10970)

mindest committed 4 years ago

Verified 3c5853dc

amdmigraphx_ep-add ops to be supported by migraphx and fixed a bug in check ops to be supported (#10496)

scxiao committed 4 years ago

Verified 7ee52fb8

Fix type constraints in registration of DequantizeLinear (#10986)

Adrian Tsai committed 4 years ago

Verified ae08f966

Set the default for the STFT onesided attribute to 1, which tests expect (#10984)

smk2007 committed 4 years ago

Verified 938f3857

Fixed macros for graph transformer registration. (#10983)

chandru-r committed 4 years ago

Verified 07201726

Custom parameters (#10964)

oliviajain committed 4 years ago

Verified de384805

use #include <hiprand/hiprand.h>, not deprecated #include <hiprand.h> (#10966)

jeffdaily committed 4 years ago

Verified 9a3be9b4

fix coverage report error in master build (#10969)

mszhanyi committed 4 years ago

Verified 0efbe922

Update training packages to Pytorch 1.11.0 (#10851)

raviskolli committed 4 years ago

Verified 480c7931

Support ORT WASM compilation with the training flag (#10973)

baijumeswani committed 4 years ago

Verified 565318ce

Ignore DequantizeLinear nodes in CommonSubexpressionElimination optimizer (#10934)

skottmckay committed 4 years ago

Verified b28e5064

Adds missing numpy type when looking for the ort correspondance (#10943)

xadupre committed 4 years ago

Verified b88fb68f

add build flag for emscripten settings (#10963)

fs-eire committed 4 years ago

Verified dce5d719

Add multi-dim dft test, and fix complex idft (#10947)

smk2007 committed 4 years ago

Verified 027565b3

allows multiple '--cmake_extra_defines' flags (#10953)

fs-eire committed 4 years ago

Verified 2da82fd0

Newer Older