microsoft/onnxruntime

Pull Requests Commits

Check for null input OrtValue

Pranav Sharma committed 3 years ago

24225e11

[CUDA] Split/Concat Kernel Optimization (#12175)

Lafi7e committed 3 years ago

Verified 173bcdbc

[js/web] use windowed Chrome for perf mode (#12157)

fs-eire committed 3 years ago

Verified ced7c2de

Add --disable_shape_inference option to optimizer.py (#12215)

tianleiwu committed 3 years ago

Verified b81b6526

Fix bug where onnxruntime_USE_NCCL flag would default to ON (#12195)

seanmurr1 committed 3 years ago

Verified 93229949

remove identity in transformers model graph fusion (#12194)

tianleiwu committed 3 years ago

Verified 17b84c78

Add file mapping for windows platform. (#12183)

caoting-dotcom committed 3 years ago

Verified 4d38b84e

remove wrong placed libs (#12201)

leqiao-1 committed 3 years ago

Verified 09af4a7f

[TVM EP][CI] Integrate TVM EP into ORT public CI on Windows (#12161)

KJlaccHoeUM9l committed 3 years ago

Verified d31db1aa

Fix line spacing/break issue, extend existing tests (#12191)

msftlincoln committed 3 years ago

Verified 52095fb0

OnnxRuntime Eager: Implement log_softmax with ONNX Ops (#12190)

msftlincoln committed 3 years ago

Verified a2dc6d32

bitwise_and ONNX support (#12189)

msftlincoln committed 3 years ago

Verified 9bca8405

Simple eager training models (#12180)

WilBrady committed 3 years ago

Verified 89bf6c9b

add comment to explain local scalar dense (#12179)

msftlincoln committed 3 years ago

Verified fafb2414

fixing positions for beam search gpt2 (#12156)

viboga committed 3 years ago

Verified 05c31a03

Update eager Readme.md (#12170)

WilBrady committed 3 years ago

Verified 9ebef91a

[UPDATE] update AMD CI pipeline to Rocm5.2 with torch1.11 (#12162)

PeixuanZuo committed 3 years ago

Verified 7b53b223

Remove Apex Dependency For Deepspeed FP16_Optimizer (#12077)

Lafi7e committed 3 years ago

Verified a7eb9fe3

Eager mode: Fix some python warnings. (#12167)

WilBrady committed 3 years ago

Verified 5da1e5d3

ppc64le: Optimizing the MlasQLinearMulKernel() to use VSX instructions (#12051)

maxiwell committed 3 years ago

Verified 51f8456c

x86/64 U8S8 Gemm Precision Fix (#12088)

chenfucn committed 3 years ago

Verified 040c2f45

Fix NonZero eager impl. (#12143)

WilBrady committed 3 years ago

Verified 48647bc7

[TVM EP] support build on Windows (#11851)

vvchernov committed 3 years ago

Verified 3b0aaa9e

Fix GH issue 12151 by using inverse perms for updating DQ axis attribute (#12158)

skottmckay committed 3 years ago

Verified 75cf5dc2

Rework cmake for kernel_explorer (#12079)

cloudhan committed 3 years ago

Verified 785f7497

[add] Add operator gemmfastgelu for ROCM (#12101)

PeixuanZuo committed 3 years ago

Verified 5579d81f

Use updated symbolic_helper.check_training_mode (#11900)

jingyanwangms committed 3 years ago

Verified a9d0d332

List 3.10 as supported python version and remove 3.6 (#12141)

RandySheriffH committed 3 years ago

Verified 178a413c

[java] First part of the JNI error handling rewrite (#12013)

Craigacp committed 3 years ago

Verified e0ed9f0f

Eager mode generator improvements for multiple onnx operators and extra test cases (#12111)

msftlincoln committed 3 years ago

Verified a6fd1a3b

Older