saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-06-02 23:39:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
1707 commits 2856 branches 57 tags 1.3 GiB
Commit graph

1707 commits

This branch
This branch
All branches
Author SHA1 Message Date
Jeff
b38b36a941
Increase max_num_graph_transformation_steps from 5 to 10 (#2633) 2019-12-16 14:17:37 -08:00
Jeff
9dddabf237
Miscellaneous DML EP fixes (#2634) 
* Misc DML EP fixes

* Fix white space

* Build fix
 2019-12-16 14:16:10 -08:00
Tracy Sharpe
c907881852
MLAS: optimize QuantizeLinear (#2660) 
This adds MLAS support to quantize a buffer using vector instructions for x86/x64/ARM64. This helper is now used for the QuantizeLinear and DynamicQuantizeLinear ops.
 2019-12-16 10:56:23 -08:00
Hector Li
47503ec7a6
Initiate the build scripts for ARM ACL (#2652) 
1. Add scripts to build Yocto image & toolchain
2. Update docker build scripts to support Onnxruntime build with ARM ACL 19.02/19.05
 2019-12-16 09:44:19 -08:00
Yang Chen
f7412899a1
added cache version for nuphar JIT binaries (#2646) 
* added cache version for nuphar JIT binaries

Previously, when the user wrongfully loaded a JIT binary generated
from a Nuphar version different from the current used one, she
would get mysterious runtime failures, because we didn't perform
any version check on JIT binaries.

This change added cache versions to the Nuphar runtime and
JIT binaries. The Nuphar runtime will issue verbose message that
informs the user version-mismatch errors.

* address CR feedback

* include NUPHAR_CACHE_VERSION in python wheel
 2019-12-14 22:46:30 -08:00
Dmitri Smirnov
7c87070b24
Import Featurizers (#2643) 
Import FeaturizerLibrary as ExternalPorject which is optional and is not registered as git submodule.
 2019-12-13 16:07:12 -08:00
Changming Sun
a46a28b7d8
Windows CI changes(#2650) 2019-12-13 12:23:49 -08:00
Tianlei Wu
18bdde34dc
Fix memory exception in Layer Norm Fusion (#2644) 2019-12-13 11:13:55 -08:00
Zhang Lei
1996129ddf
Improve performance of resize() in Nearest mode (#2626) 
Special treatment for 2D, check same size as input image.
And in 2d kernel, template use_expolation.
 2019-12-13 09:43:13 -08:00
Hector Li
6e68007d2d
fix build error for ARM (#2648) 2019-12-13 09:21:03 -08:00
Zhang Lei
fff1ed9bfc
Optimize cuda scatter() on 2D compatible. (#2628) 
* Optimize cuda scatter() on 2D compatible.

* Add some comments.
 2019-12-12 16:38:06 -08:00
Zhang Lei
e31be23c42
Cuda pad optimize when no padding is needed. (#2625) 
* Shortcut cuda Pad() when no padding is needed.
 2019-12-12 16:36:46 -08:00
Zhang Lei
8631b70c73
Improve cuda expand() opeator's performance. (#2624) 2019-12-12 16:36:06 -08:00
Jeff
ac08b58867
Add shape inference to ConvTransposeWithDynamicPads schema (#2632) 2019-12-12 14:46:06 -08:00
shahasad
4dbf9442cc
removed unnecessary batch file and fix path (#2640) 2019-12-12 14:21:02 -08:00
liuziyue
c7cd336917 epsilon attribute for layernormalization fusion (#2639) 2019-12-12 10:58:02 -08:00
Yufeng Li
d6f33dceb1 fix float16 comparison in initializer (#2629) 2019-12-11 19:03:57 -08:00
Tianlei Wu
c04647b952
Improve Embed Layer Norm Fusion for SQuAD with static input shape (#2621) 2019-12-11 15:25:35 -08:00
Zhang Lei
6859d92e2b Make sure fenced tensor could not reuse other tensor. (#2561) 
Fix random error caused by this.
 2019-12-11 12:04:54 -08:00
Hariharan Seshadri
b2d65b447f
Remove unnecessary parameter in some places in GatherElements implementation (#2612) 
* Remove unnecessary parameter in some places

* Update

* Update
 2019-12-11 11:15:12 -08:00
Eric Cousineau
8729784635 Allow providers to be set for InferenceSession at construction (#2606) 2019-12-11 10:24:43 -08:00
liuziyue
1ee250def8 EmbedLayerNormalization Fusion For Dynamic Squad Model Opset 10 (#2613) 
Support subgraph of SQuAD model exported from pytorch with dynamic input axes
 2019-12-10 23:40:55 -08:00
Yang Chen
2ca9733cee
Dump subgraph ID and fused graph ID (#2607) 
* Dump subgraph ID and fused graph ID

Dump subgraph ID and fused graph ID for better debugging

* Remove local static fused_count

added a field global_fused_count_ to NupharExecutionProvider class
 2019-12-10 19:56:39 -08:00
KeDengMS
45babd6c00
symbolic shape inference: fix warnings in GPT-2 model (#2608) 
And revise nuphar perf test on BERT squad
 2019-12-10 16:05:00 -08:00
Tianlei Wu
bc89eccb21
Support opset 11 subgraph of Squad model in Embed Layer Normalization (#2605) 
Support opset 11 Squad model that is exported from PyTorch nightly. The embed layer uses Range op which is missed in the transformer.
 2019-12-10 15:22:19 -08:00
Dmitri Smirnov
796948c6ae
Rename automl python tools folder to featurizer_ops. (#2593) 2019-12-10 10:58:40 -08:00
Tianlei Wu
6858f0a06a
Add support for opset 11 in reshape fusion (#2592) 
Support opset verion 11 in reshape fusion
 2019-12-10 10:29:41 -08:00
Adam Pocock
35ceb1a6a6 Java API for onnxruntime (#2215) 2019-12-10 08:28:46 -08:00
Yang Chen
b0128a4843
Fixed an issue in updating realized dims (#2597) 
when we update realized dims for scan's output, the sliced axis also
needs to be inclusive, i.e. we should check with "dim >= insert_inclusive_axis",
because the offset in the symbols are based on Scan sugraph.
Otherwise, we would end up with shape mismatch later.
 2019-12-09 22:56:47 -08:00
Ashwini Khade
78099701b4
Add missig env variables for mac pipeline test (#2595) 2019-12-09 21:10:43 -08:00
shahasad
41fc820f76
add path to build dir before test run (#2590) 2019-12-09 18:58:55 -08:00
Changming Sun
6e08efa6a2 Fix lto bug for protobuf and ubuntu 2019-12-09 17:34:06 -08:00
daquexian
62de8fa841 Update docs for Android NNAPI EP (#2586) 2019-12-09 14:37:03 -08:00
Hector Li
0ab54521f4
Temporarily exclude vgg19 test from Python backend test 
1. temporarily exclude vgg19 test which comsumes too much memory, run out of memory on Upsquared device. Single test pass for vgg19, need furture investigation (#2588)
2. Update docker file to decrease the docker image size
 2019-12-09 12:25:46 -08:00
Ryan Hill
36eb1771ba
Update version (#2584) 2019-12-08 18:00:12 -08:00
liuziyue
200f4b4ea6 EmbedLayerNormalization Fusion Improvement (#2553) 
Embedding layer norm fusion improvements - add more checks
 2019-12-07 23:14:26 -08:00
KeDengMS
0f12346d76
[Nuphar EP] fixes for some object detection models (#2581) 
Update notebook tutorial with multi-threaded int8 GEMM from #2517
 2019-12-07 13:37:00 -08:00
Ryan Hill
cbc398bb75
Ryanunderhill/packagename test (#2582) 2019-12-07 12:08:46 -08:00
Ashwini Khade
c06dbd8311 Add ConvTranspose1D (#2578) 2019-12-07 08:50:02 -08:00
Mark
79847f39b3 Fix file not found error during docker build. (#2569) 2019-12-07 08:49:47 -08:00
Yufeng Li
5575766a53
Add more check on SkipLayerNorm and BiasGelu fusion (#2574) 2019-12-06 15:36:02 -08:00
Changming Sun
262ee9dc5a Fix a warning found in the latest VS release 2019-12-06 15:07:21 -08:00
Yufeng Li
34beafc51c
make layernorm fusion to support opset 11 (#2545) 2019-12-06 13:06:36 -08:00
shahasad
eeb28a80c0
setup java ci mac (#2570) 2019-12-06 11:43:40 -08:00
Tianlei Wu
038ee91da5
Allow sequence length to be symbolic (#2559) 2019-12-06 10:13:56 -08:00
George Wu
73c682b97c
disable onnx_test_runner -x invocations for dnnl (#2568) 2019-12-05 23:05:34 -08:00
Changming Sun
7eddac16c2
Re-enable Windows C# tests (#2564) 2019-12-05 21:22:31 -08:00
Ryan Hill
854362cf05
Update win-x86-ci.yml (#2557) 
Fix build pipeline break
 2019-12-05 18:44:12 -08:00
Changming Sun
ace132f9aa
Fix android build (#2558) 2019-12-05 15:03:22 -08:00
Sreekanth Yalachigere
4c996a8699 DNNL CMAKE update (#2548) 2019-12-05 13:48:57 -08:00