IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
12,130 Commits 38 Branches 54 Tags
3e0f529700dc82d01658dbf55b19604048475311
Commit Graph
2377 Commits
Author SHA1 Message Date
Jade Cho 2394732055 [dGPU] Use cldnn concat if data type is fp32. (#19240) 2023-08-17 09:14:26 +00:00
Min, Byungil 1b9de79d0d [GPU] Resolve fp16 overflow of mul (#19173) 
+ Fixed black output image by Nan output.
+ Resolved fp16 overflow of gemm primitive before softmax
+ Added fused post ops of clamp activation not to get inf which caused Nan output
+ Added new pass modify_fused_ops

Signed-off-by: Min, Byungil <byungil.min@intel.com>
 2023-08-17 18:09:10 +09:00
Andrei Gorbachev 49bbcb4cf6 [ПЗГ] fix coverity (#19204) 2023-08-17 12:46:35 +04:00
Wilson Seok b656feee57 [GPU] fix priorbox legacy code in CreateOp() and allow to use max_size is null (#19143) 2023-08-16 17:54:04 +04:00
Min, Byungil ef6c8c1d66 [GPU] Make gemm_tiled_opt support outer axis (#19210) 2023-08-16 16:43:47 +04:00
Maksim Kutakov 74c237de2e [CPU] Remove skip for the first inPlace base edge (#19185) 2023-08-16 11:49:04 +04:00
Maksim Kutakov d13dff06f3 [CPU] Disable gather inPlace for out of range indices (#19159) 2023-08-16 10:32:31 +04:00
Andrew Kwangwoong Park 8a0a4df941 [GPU] Fix unintentional expansion of pad's rank into 2D (#19106) 
* [GPU] Fix unintentional expansion of pad's rank into 2D

* Add explicit 1d pad TCs for ov_gpu_unit_tests
 2023-08-16 14:29:00 +09:00
Roman Lyamin 188434e969 [GPU] image_2d_weights_winograd_6x3_s1 formats fix (#19205) 2023-08-16 08:59:43 +04:00
yanlan song 13f8ff4a40 fix rt_info meta parse race condition (#19196) 
* fix a possible race condition

Signed-off-by: fishbell <bell.song@intel.com>

* clang

Signed-off-by: fishbell <bell.song@intel.com>

* reimpl

Signed-off-by: fishbell <bell.song@intel.com>

* add cases

Signed-off-by: fishbell <bell.song@intel.com>

* implement test in 2.0

Signed-off-by: fishbell <bell.song@intel.com>

* clean up unchanged files

Signed-off-by: fishbell <bell.song@intel.com>

* add pragma once

Signed-off-by: fishbell <bell.song@intel.com>

---------

Signed-off-by: fishbell <bell.song@intel.com>
 2023-08-15 15:33:01 +00:00
Sergey Shlyapnikov d13ae31a61 [GPU] Allow to use infinity value as a Pad's fill value (#19201) 2023-08-15 14:31:49 +00:00
Gorokhov Dmitriy 5ff67fca40 [CPU][ARM] Fixed execution mode behavior (#19192) 2023-08-15 16:46:05 +04:00
Eddy Kim 0194cdce3b [GPU] updated to add reorder after parameter when it is connected to quantize (#19182) 
* updated to add reorder after parameter when it is connected to quantize

* removed redundant code
 2023-08-14 23:29:49 -07:00
Egor Duplenskii 37a6848f26 [CPU][ARM][FP16] Prefer Convert over Reorder for conversion (#19148) 2023-08-15 09:11:50 +04:00
Roman Lyamin 9be28328bf Added giy_xs_os_xsv2_osv8__ao32 format to to_weights_layout map (#19181) 2023-08-15 08:49:50 +04:00
Paul Youngsoo Ahn e2db808495 [GPU] Fix multiple output issue in get_output_layout(#19186) (#19186) 2023-08-14 20:49:21 +00:00
Aleksandr Voron e48b2dfc34 [CPU][ARM] Set FP16 inference precision by default for non-convolution networks on ARM (#19069) 2023-08-14 22:22:39 +04:00
Egor Duplenskii 680333b2db [CPU][ARM] Enable FP16 precision for CumSum node (#19176) 2023-08-14 15:22:43 +00:00
Anton Voronov 87f9b2bdf8 [CPU] Fixed checks in snippets_mark_skipped for MatMul weights (#19141) 2023-08-14 16:04:35 +04:00
Sergey Shlyapnikov 86c4c6785d [GPU] Don't invalidate users after reorder insertion (#19137) 2023-08-14 14:16:54 +04:00
Maksim Kutakov 4996d1f034 TensorMemoryBlob: sync data handle on alloc call (#19149) 2023-08-14 13:22:11 +04:00
Egor Duplenskii f09d2e2666 [CPU][ARM] Perf fixes for FP16 precision (#18973) 2023-08-14 09:22:03 +04:00
Mingyu Kim 4e96b6ba9d [GPU] Do not select onednn format for asymmetric weight (#19140) 
OneDNN does not support asymmetric weight.
 2023-08-14 12:59:07 +09:00
Ilya Lavrenov 6067ab17ba Aligned target and files (libraries) names (#19161) 2023-08-13 11:31:44 +04:00
Wanglei Shen 0368474c8a fix manually setting Pcore+Ecore manually when model prefers Pcore only (#19160) 2023-08-12 18:19:02 +08:00
Gorokhov Dmitriy 2bdb7bd23e [CPU] FullyConnected acceleration with 8bit weights decompression on SPR (#19111) 2023-08-11 14:14:32 +04:00
Ilya Lavrenov a251346031 [CPU] Give different output name for dnnl (#19136) 
* Give different output name for dnnl

* Apply suggestions from code review

added prefix to library name
 2023-08-11 13:54:44 +04:00
Ilya Lavrenov cc5ab26a95 Remove explicit CRT options from MLAS (#19115) 2023-08-11 13:03:12 +04:00
Mingyu Kim c1c4c4cd51 [GPU] Onednn version update to latest 3.2 (#19135) 2023-08-11 17:41:41 +09:00
Anton Voronov 2a3132941c [FIX][CPU] Convert CF: added convert childs number check (#19022) 
* [FIX][CPU] Convert CF: added convert childs number check

* code style fix

* WIP: [CPU][TESTS] Convert CF: Added subgraph test with two outputs for Convert node

* MatMulDecompressConvertTest refactoring + instances fixes

* removed KeepConstAndDecompressionForMatMul pass and added cpu callback for KeepConstAndDecompression

* MatMulDecompressConvertTest2: added graphs and small problem description

* small review fix
 2023-08-11 12:10:00 +04:00
Sun Xiaoxia 0ed1749ad9 [CPU] Removed socket id=-1 workaround on macOS (#18779) 2023-08-11 10:38:21 +04:00
Wanglei Shen 2ed5e50bd1 [CPU] Update stream info for mixing MAIN_CORE_PROC and HYPER_THREADING_PROC (#18863) 2023-08-11 10:31:31 +04:00
Tomasz Adamowicz d6e94212b8 Temporary deactivation of sporadically failing tests (#19118) 2023-08-10 17:54:51 +04:00
Vladimir Paramuzov cde0893271 [GPU] Use ov::Tensor intead of ngraph::HostTensor (#18925) 2023-08-10 09:43:45 +00:00
Zhang Yi 7d192c9239 Update mlas with i32 build fix (#19107) 2023-08-10 13:38:16 +04:00
Eddy Kim 9deef1480a updated to init onednn attr (#19055) 2023-08-10 07:38:32 +00:00
Sergey Shlyapnikov 32a6a31de2 [GPU] Prevent Conv's input data type changing at reorder_inputs pass (#19042) 2023-08-10 10:59:32 +04:00
Sergey Shlyapnikov d91d72c89c [GPU] RandomUniform new shape inference for dynamism support (#19087) 2023-08-10 09:43:08 +04:00
hyunback kim f683fabcbf [GPU] Add Disable Winograd convolution debug config. (#18701) 
Signed-off-by: hyunback <hyunback.kim@intel.com>
 2023-08-10 10:58:57 +09:00
Roman Lyamin 575f497c8b [GPU] Fix data race in CompilationContext (#19103) 2023-08-10 00:50:49 +00:00
Stefania HerganeandDan 24f8c4105e [EISW-89824] [master] Rename VPUX to NPU (#19004) 
* Change `VPUX`/`VPU` occurrences to `NPU`

* Switch `HARDWARE_AWARE_IGNORED_PATTERNS` VPU to NPU

* Rename `MYRIAD plugin`

* Rename vpu_patterns to npu_patterns in tools/pot

* Rename vpu.json to npu.json in tools/pot

* Rename restrict_for_vpu to restrict_for_npu in tools/pot

* Change keembayOptimalBatchNum to npuOptimalBatchNum

---------

Co-authored-by: Dan <mircea-aurelian.dan@intel.com>
 2023-08-10 00:20:07 +04:00
Nadezhda Ageeva 81645aaeda [HETERO] Avoid unexpected dynamism (#18862) 
* [HETERO] Add ConstantFolding in compile modelto avoid unexpected dynamism after model split. Add new property, which shows number of subgraphs

* Remove check for dynamic subgraph
 2023-08-09 17:22:20 +04:00
Aleksandr Voron 19b3f062af [CPU] Replace memory_order_relaxed with release/acquire in UpdateNodesBase class (#19028) 2023-08-09 13:28:44 +04:00
Roman Lyamin 2d5a4e20ec [GPU] Fix premature termination of async compilation task in tests (#19035) 2023-08-09 12:47:06 +04:00
Roman Lyamin 7d89c93ded [GPU] Added gi_yxs_os_yxsv2_osv16 format to to_weights_layout map (#19074) 2023-08-09 12:03:48 +04:00
Andrew Kwangwoong Park df659daf44 [GPU] Add reorder if the input format of eltwise is different from the output format (#19068) 
* [GPU] Add reorder if the input format of eltwise is different from the output format

* Fix failed TCs for eltwise's mixed layout
 2023-08-08 23:21:54 -07:00
Kelvin Choi e60da26894 [GPU] Support dynamic BatchToSpace and SpaceToBatch (#18630) 2023-08-09 05:56:55 +00:00
Wang, Yang 5db7579bea [AUTO Plugin] Update document for testing AUTO plugin from benchmark app (#18809) 
* Update guide document to show the detailed steps of testing AUTO plugin from benchmark app
 2023-08-09 13:37:06 +08:00
David Nam a8f4e114f3 [GPU] Check if padding is supported for static shape when buffer fusing (#18861) 
* [GPU] Check if padding is supported even for static shape when buffer fusing

* Add unit test

* Remove unnecessary logic
 2023-08-08 22:15:56 -07:00
hyunback kim 58bd8e7f0b [GPU] Add disable winograd convolution in execute_config. (#18964) 
* [GPU] Add disable winograd convolution in execute_config.

Signed-off-by: hyunback <hyunback.kim@intel.com>
 2023-08-09 13:32:36 +09:00