microsoft/onnxruntime

Pull Requests Commits

RandySheriffH committed 4 years ago

b7f4a866

RandySheriffH committed 4 years ago

b819ed7b

add workspace func

RandySheriffH committed 4 years ago

1b8a4a5e

RandySheriffH committed 4 years ago

56f36703

reset alpha beta

RandySheriffH committed 4 years ago

7babe68f

RandySheriffH committed 4 years ago

22fa25ff

handle status error from cudnn 8.x

RandySheriffH committed 4 years ago

94559947

exclude test from cpu

RandySheriffH committed 5 years ago

bb7949f8

RandySheriffH committed 5 years ago

23651efc

update quantize to support basic optimization and e2e example for image classification (#6313)

yufenglee committed 5 years ago

Verified c24f2950

Fix DerefNullPtr issues raised by SDLNativeRules. (#6348)

Pranav Sharma committed 5 years ago

Verified 5b9d993a

Train BERT Using BFloat16 on A100 (#6090)

Lafi7e committed 5 years ago

Verified 4df356d1

fix the pipeline failure (#6346)

guoyu-wang committed 5 years ago

Verified e35db194

Add support for running Android emulator from build.py on Windows. (#6317)

edgchen1 committed 5 years ago

Verified 042053c5

[NNAPI] Add pow support (#6310)

guoyu-wang committed 5 years ago

Verified b220feee

remove gemmlowp submodule (#6341)

tracysh committed 5 years ago

Verified fcd9fc9b

Remove OpSchema dummy definition. Only needed for Function now, and we can just exclude the method in Function (#6321)

skottmckay committed 5 years ago

Verified cfd6f100

remove needless print out

RandySheriffH committed 5 years ago

0e8e8347

changed wording. (#6337)

Tixxx committed 5 years ago

Verified d367941c

add external data support to tensor proto utils (#6257)

askhade committed 5 years ago

Verified f7034b9b

Enable add + softmax fusion for Rocm platform (#6259)

Suffian Khan committed 5 years ago

Verified 62e40459

Delete float16.py (#6336)

oliviajain committed 5 years ago

Verified 56ab2166

optimize cuda conv by fused activation

RandySheriffH committed 5 years ago

ff2be1e3

MLAS: add fallback implementation for quantized GEMM (#6335)

tracysh committed 5 years ago

Verified 87ec1f62

Use onnxruntime_USE_FULL_PROTOBUF=OFF for the cuda execution provider (#6340)

alberto-magni committed 5 years ago

Verified 5623cc6d

Liqun/enable pipeline parallel test (#6331)

liqunfu committed 5 years ago

Verified aeca96ca

Quantization support for split operator with its NHWC support (#6107)

zhanghuanrong committed 5 years ago

Verified f77ff1bc

Java: add Semmle to Java publishing pipelines (#6326)

yuslepukhin committed 5 years ago

Verified 6b73bae0

Remove false positive prefast warning from threadpool (#6324)

tlh20 committed 5 years ago

Verified aacc8dbf

fix opset imports for function body (#6287)

askhade committed 5 years ago

Verified 0ed56d49

Older