microsoft/onnxruntime

Pull Requests Commits

Add CUDA Scan operator. (#2403)

skottmckay committed 6 years ago

Verified 87d3794a

Adjust codegen vectorization width from target (#2439)

baowenlei committed 6 years ago

Verified 9b7b5e2c

add Im2col<uint8_t> (#2438)

tracysh committed 6 years ago

Verified a5daa6fa

Fix the type constraints on CUDA If operator to exclude strings. (#2431)

skottmckay committed 6 years ago

Verified 3be554c2

Improved documentation for onnxruntime::utils::SwapByteOrderCopy(), added precondition check.

edgchen1 committed 6 years ago

8647201a

Add CUDA If operator. (#2377)

skottmckay committed 6 years ago

Verified be12cdc7

Added support for Pad-2 operator in OpenVINO-EP (#2405)

psfoley committed 6 years ago

1cb6bdc3

onnxrt server documentation update (#2396)

avidiyal committed 6 years ago

95e8c337

feat(treeregressor): Update TreeEnsembleRegressor for type support (#2389)

groszewn committed 6 years ago

0e947bd3

Fix the issue in matmul_add_fusion (#2407)

HectorSVC committed 6 years ago

Verified 367361fc

[NupharEP] Update notebook and docker image (#2416)

KeDengMS committed 6 years ago

aa7c79ea

Add Reshape Fusion (#2395)

tianleiwu committed 6 years ago

Verified e1c17fd1

Minor optimization: if a node has already been placed, there's no need to find a kernel for it. (#2417)

Pranav Sharma committed 6 years ago

Verified f268e69c

fix cross compile bug (#2415)

baowenlei committed 6 years ago

5ab7041f

[NupharEP] force some low/zero cost ops to be inlined (#2409)

KeDengMS committed 6 years ago

Verified 1e03ce84

Add opset 11 versions of the existing CUDA operators that had negative axis support explicitly added. (#2398)

skottmckay committed 6 years ago

Verified c1d757a0

Change CUDA implementation of Transpose to support all fixed size tensor types (#2387)

skottmckay committed 6 years ago

Verified 6cc57721

Avoid using the default logger in the graph lib and optimizers (#2361)

Changming Sun committed 6 years ago

Verified 109b3cb4

[NupharEP] Multiple optimizations (#2380)

KeDengMS committed 6 years ago

Verified b15e43a5

Fix reuse logic in allocation planner. (#2393)

Pranav Sharma committed 6 years ago

Verified 7e164eaa

Fixed compilation with ngraph (#2388)

ilya-lavrenov committed 6 years ago

b90d55b7

fix BUILD.md typo (#2375)

nihui committed 6 years ago

dde410e0

Another try to stabilize CUDA CI (#2383)

KeDengMS committed 6 years ago

Verified 51571030

Skip layer norm transform (#2350)

liuziyue committed 6 years ago

Verified ffa28125

Fuse Add + Gelu (#2360)

yufenglee committed 6 years ago

Verified 8ed2928d

Layer Norm Fusion Fix (#2379)

liuziyue committed 6 years ago

Verified 4b72fedb

Add opset 11 version of Split to CUDA ops (#2376)

skottmckay committed 6 years ago

Verified 8c733c8d

Fix bug with Slice. Need to pass in flattened input dimensions so the initial offset into the input is calculated correctly. (#2372)

skottmckay committed 6 years ago

Verified c0d23d5f

Extend OneHot CPU kernel to support more types (#2311)

KeDengMS committed 6 years ago

Verified 9e26f4de

update output size calculation for resize (#2366)

askhade committed 6 years ago

Verified 437772d5

Newer Older