microsoft/onnxruntime

Pull Requests Commits

Introduce ORTModule training API to ONNX Runtime

Thiago Crepaldi committed 5 years ago

89d45069

Update onnxruntime_perf_test.exe to accept free dimension overrides (#6962)

Ori Levari committed 5 years ago

Verified 9f84819f

fixed type to experimental session constructor (#6950)

dmedine committed 5 years ago

Verified f723ff22

Attention fusion detect num_heads and hidden_size automatically (#6920)

tianleiwu committed 5 years ago

Verified 4884eee6

Add *args support for ORTModule inputs (#6883)

Sergii Dymchenko committed 5 years ago

Verified ce403eea

Implement QLinearAveragePool with unit tests. (#6896)

zhanghuanrong committed 5 years ago

Verified acfe7ac4

Support ROCM EP for ORTModule (#6967)

weixingzhang committed 5 years ago

Verified 1e13e266

MLAS: quantized GEMM update (#6916)

tracysh committed 5 years ago

Verified a8b897f7

Fix warning from setting multiple MSVC warning level options. (#6917)

edgchen1 committed 5 years ago

Verified bc319bd7

Use DLPack for Graph Inputs and External Outputs of YieldOp (#6968)

Lafi7e committed 5 years ago

Verified 8468099f

Relax atol for some ORTModule UTs (#6969)

Lafi7e committed 5 years ago

Verified 3f579fac

Enable type reduction in EyeLike, Mod, random.cc CPU kernels. (#6960)

edgchen1 committed 5 years ago

Verified d5ed3e7f

fix stream sync issue (#6954)

tianleiwu committed 5 years ago

Verified 89916fdb

Update baseline due to loss scale fix (#6948)

wschin committed 5 years ago

Verified bdaea1d9

Fix broken link in server usage and remove absolute path from dockerfiles readme (#6926)

0xRaduan committed 5 years ago

Verified 743a93fa

Support ORTModule on ROCm EP (#6945)

weixingzhang committed 5 years ago

Verified 534adbb0

Add UT correctness and address comments for previous symbolic shape PR (#6930)

ytaous committed 5 years ago

Verified 3b2847b2

Add GPU support for DNNL endpoint (#6741)

georgen117 committed 5 years ago

Verified ba51774a

Clean ORTModule dev branch (#6944)

Thiago Crepaldi committed 5 years ago

Verified 5303b33f

Interchange Cast and Transpose operations to facilitate Transpose-MatMul fusion (#6924)

satyajandhyala committed 5 years ago

Verified 48eebed8

Add UseCount for External Outputs (#6894)

Lafi7e committed 5 years ago

Verified 91c6a330

Support parsing an array of values stored as an attribute in a custom op (#6878)

hariharans29 committed 5 years ago

Verified c8e2e319

Enable build with bitcode for iOS (#6905)

guoyu-wang committed 5 years ago

Verified e64eff1f

Rename op kernel type control 'supported types' to 'default types'. (#6886)

edgchen1 committed 5 years ago

Verified 73fe1f2d

Assert that the data is on the same device as ORTModule (#6942)

baijumeswani committed 5 years ago

Verified f1ade14e

Only set _native folder for Microsoft.AI.MachineLearning package (#6939)

smk2007 committed 5 years ago

Verified 67c67408

MLAS: workaround LLVM x86 assembler (#6922)

tracysh committed 5 years ago

Verified bc276521

Add tests of Attention and QAttention for pruned model (#6914)

tianleiwu committed 5 years ago

Verified b89f52c2

Fix broken Java API link (#6826)

dennyac committed 5 years ago

Verified f2f60eed

Enable type reduction for Clip, MaxPool, and Pad CPU kernels. (#6918)

edgchen1 committed 5 years ago

Verified 15d81fb6

Newer Older