microsoft/onnxruntime

Pull Requests Commits

Address comments from ORTModule master merge (#7101)

SherlockNoMad committed 4 years ago

Verified ab86634c

Add support for large models (#7113)

Thiago Crepaldi committed 4 years ago

Verified a01f1519

icnrease timeout (#7145)

Suffian Khan committed 4 years ago

Verified 2b31b80b

update bert quantization notebook (#7137)

yufenglee committed 4 years ago

Verified 3771e0bf

Disable MatmulTransposeFusion for CPU EP (#7135)

KeDengMS committed 4 years ago

Verified c9b29fbd

Fix bugs in sparsify script (#7134)

yuslepukhin committed 4 years ago

Verified 2bf54bca

Add function-body to SoftmaxGrad (#6988)

gramalingam committed 4 years ago

Verified cc0e7bee

Add session option configuration to enable GeluApproximation (#7131)

tianleiwu committed 4 years ago

Verified 53c123dc

QDQ implementation (#7033)

yufenglee committed 4 years ago

Verified 8e54b76e

Exclude profiler from minimal build (#7115)

RandySheriffH committed 4 years ago

Verified 865c6761

Add New AllocKind for YieldOp Outputs, Run YieldOp with InferenceSession in UT (#7125)

Lafi7e committed 4 years ago

Verified fda04706

Promote BiasDropout from orttraining to onnxruntime (#7116)

SherlockNoMad committed 4 years ago

Verified 1c8d8744

Move IOBinding and RunOptions to ctx (#7028)

jingyanwangms committed 4 years ago

Verified cd67f12a

Move TensorRT Windows CI build to the machine pool (#7127)

Changming Sun committed 4 years ago

Verified 2e3bbad1

[NNAPI EP] Fix error for QLinearAdd with an initializer as input (#7093)

guoyu-wang committed 4 years ago

Verified 1c04eec2

Deepspeed pipeline parallel and fairscale sharded optimizer test samples with ORTModule (#7078)

harshithapv committed 4 years ago

Verified 540eac25

Add missing Python dependencies for ORT training (#7104)

KeDengMS committed 4 years ago

Verified 6987106b

Fix a bug in quant GEMM and add an unit test (#7111)

yufenglee committed 4 years ago

Verified fffe16cb

Delete an unused file: download_test_data.py (#7109)

Changming Sun committed 4 years ago

Verified b07e168a

update Dockerfile for workaround for issue in RCCL for rocm4.0 (#7108)

Suffian Khan committed 4 years ago

Verified 5cb89344

Update ORTTrainer to permit Rocm and permit export of opset 13 (#7059)

Suffian Khan committed 4 years ago

Verified c0994fdf

Enable type reduction for Shrink, Sign, SplitToSequence CPU kernels (#7090)

edgchen1 committed 4 years ago

Verified 53392664

Support for models having partially non trainable parameters (#7058)

baijumeswani committed 4 years ago

Verified c3310efd

Pass arguments to azure_scale_set_vm_mount_test_data from perf test ci pipeline (#7094)

baijumeswani committed 4 years ago

Verified a7a2a16e

fix a bug in global average pool and add unit test (#6913)

yufenglee committed 4 years ago

Verified c965878a

cmake: support install target with generated pkg-config file (#7076)

Aaron Boxer committed 4 years ago

Verified 230c1374

upload ort-gpu-training python nightly package to azure feed (#6998)

liqunfu committed 4 years ago

Verified 309885b0

MLAS: add 32-bit transpose support (#7092)

tracysh committed 4 years ago

Verified 416ee3c4

ORTModule support non-differentiable module output (#7048)

SherlockNoMad committed 4 years ago

Verified 5ec0e715

Make our CUDA code be compatible with the latest VS2019 update (#7062)

Changming Sun committed 4 years ago

Verified be45a59d

Newer Older