microsoft/onnxruntime

Pull Requests Commits

try multi-thread

chentaMS committed 3 years ago

d8d9119d

Merge branch 'PE' of https://github.com/microsoft/onnxruntime into PE

chentaMS committed 3 years ago

fb974be8

in memcpytohost, sync stream before invoke next kernel

chentaMS committed 3 years ago

f9a10a67

fix build break

Cheng Tang committed 3 years ago

129806d9

fix linux break

Cheng Tang committed 3 years ago

7f50c5b4

update the memory reuse planner; disable memory pattern for GPU pinned memory

chentaMS committed 3 years ago

3c733a8e

merge from master

Cheng Tang committed 3 years ago

7353dea3

fix nv profiling related build error

Cheng Tang committed 3 years ago

04abae9f

Update ORTModule Default Opset Version to 15 (#12419)

Lafi7e committed 3 years ago

Verified e85e31ee

CUDA kernel for ClipGradNorm for TensorSeq gradients (#12412)

baijumeswani committed 3 years ago

Verified a7d62907

[DELETE] delete python package rocm4.3.1 (#12480)

PeixuanZuo committed 3 years ago

Verified 3e1b0ac4

Fix Python Packaging CI (Rocm) (#12477)

ytaous committed 3 years ago

Verified b879dca5

Rework parts of Graph::Resolve to reduce memory usage (#12176)

skottmckay committed 3 years ago

Verified 8d830adf

Add composable kernel GEMM baseline for kernel explorer (#12364)

cloudhan committed 3 years ago

Verified f39354d7

[CUDA] BiasSoftmax Supporting New Pattern (#12361)

Lafi7e committed 3 years ago

Verified 37995a72

fix build breaks in trainging debug build

Cheng Tang committed 3 years ago

4cda94a9

Lironkesem/unsqueeze_and_squeeze (#12421)

LironKesem committed 3 years ago

Verified d452462b

Remove dynamic allocation for ThreadPool ParallelSection (#12429)

yuslepukhin committed 3 years ago

Verified a4ef0e7f

Enable quant op to share quantization parameter between input and ouput (#12408)

yufenglee committed 3 years ago

Verified ac10f33d

Fix merge issue (#12459)

jslhcl committed 3 years ago

Verified e25d95f3

Minor doc fixes (#12388)

RyanUnderhill committed 3 years ago

Verified 52d46997

Merge branch 'master' into PE

jslhcl committed 3 years ago

Verified 993e4375

Refactor InferenceSession Load member functions. (#12430)

edgchen1 committed 3 years ago

Verified 3efd9a73

dev notes for layout transformer (#12396)

askhade committed 3 years ago

Verified 97268e02

Update script to find optimizers that potentially need supported opset updates (#12330)

skottmckay committed 3 years ago

Verified a3de1bbf

[ROCm] Add AveragePool, GlobalAveragePool, MaxPool, GlobalMaxPool Ops (#11968)

xinyazhang committed 3 years ago

Verified 77cab7a3

[oneDNN EP] Optimized DynamicQuantizeLinear operator (#12403)

Erick Muñoz committed 3 years ago

Verified d1497bdf

fix error moving a temporary object prevents copy elision (#12442)

jslhcl committed 3 years ago

Verified f15f168b

Perform graph transformations during offline tooling (#12422)

baijumeswani committed 3 years ago

Verified 7f58bd72

Container and memory allocation guidelines (#12387)

yuslepukhin committed 3 years ago

Verified dc984a03

Older