IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
13,806 Commits 38 Branches 54 Tags
master
Commit Graph
43 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Anton Voronov d90e23fc30 [CPU] Sparse weights decompression feature: changed comp_tile_len data type from int16_t to int (#20707) 2023-10-31 09:34:27 +04:00
Anton Voronov aa6adcd261 Gather: removed indices normalization pass and added some checks for Gather-8 -> Gather-7 conversion (#19666) 
* Do not normalize negative indices for Gather v8

* code style fix

* added transformation test with accuracy check for Gather-v8

* removed GatherNegativeConstIndicesNormalize transformation at all

* ConvertGather8ToGather7 conversion: added more checks

* Introduced shared Gather8withIndicesDataLayerTest: added CPU, GPU instances

* code style fix

* small fix

* review fixes

* do negative indices normalization if possible

* code style fix

* refactor cpu test instances

* code style fix
 2023-10-10 11:17:33 +04:00
Anton Voronov c417d15432 [CPU] FullyConnected: sparse weights fix (#20117) 2023-10-10 10:24:38 +04:00
Anton Voronov 3ce48fc3d6 Fixed OV static build issues (#19976) 2023-09-26 10:23:00 +04:00
Anton Voronov 49c4c922ff [CPU][OneDNN] Fix zero pad perf issues (#19417) 2023-08-29 16:55:45 +00:00
Anton Voronov 59d58b2296 [CPU][ONEDNN] jit_uni_dw_conv_row_f32: fixed post ops start index (#19126) 2023-08-23 15:52:09 +04:00
Anton Voronov 4f29e60742 FIxed is_on_constant_path() using in all places (#19239) 
* Fixed matmul weights check in snippets_mark_skipped

* fix

* ConvertMatMulToFC: is_on_constant_path fix

* [TESTS] added SplitMatMulConcat subgraph test

* MarkDequantizationSubgraph: is_on_constant_path fix
 2023-08-18 14:01:07 +04:00
Anton Voronov 87f9b2bdf8 [CPU] Fixed checks in snippets_mark_skipped for MatMul weights (#19141) 2023-08-14 16:04:35 +04:00
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
Anton Voronov eacdc24d54 [CPU] Transpose constant folding on cpu plug-in side for MatMul op (#18877) 2023-08-07 22:17:08 +04:00
Anton Voronov 60e40843c0 [CPU] f16 constant folding on cpu plug-in side for MatMul only (#18079) 2023-07-20 08:58:46 +04:00
Anton Voronov 2547301fa7 [CPU] gemm convolution: fixed bias offset (#17357) 2023-06-07 17:15:01 +04:00
Anton Voronov 5e64a7c0c1 [DOC] cpu documentation fixes (#17815) 2023-06-01 12:26:06 +04:00
Anton Voronov 263e51a1be [DOC][CPU] Documentation update (#17784) 2023-05-31 10:37:32 +04:00
Anton Voronov d9b04be3ae [DOC] small fix for sparse weights decompression feature documentation (#17315) 2023-05-02 15:50:37 +02:00
Anton Voronov b64cbff10b [CPU] FQ shape agnostic kernel (#16585) 2023-04-03 09:55:49 +04:00
Anton Voronov 43fca3d231 [CPU] Introduced shape agnostic eltwise (#15976) 2023-03-31 11:28:54 +04:00
Anton Voronov 63e0b47a7d [DOCS][CPU] Added sparse weights decompression feature description (#14730) 
* [DOCS][CPU] Added sparse weights decompression feature description

* sparsity documentation fix

* fixed code snippet names

* moved code snippets into cpu folder

* review changes

* review fixes 2

* review fixes 3

* code style fix

* small fix
 2022-12-20 22:15:44 +01:00
Anton Voronov 3caebffa5c [CPU] Sparsity weights decompression (#13775) 2022-12-04 22:43:53 +04:00
Anton Voronov d9b1f10074 [CPU] [OneDNN] disabled unused amx primitives and conv primitives (#10326) 2022-02-18 16:36:49 +03:00
Anton Voronov 7a82bb2acb [CPU] fixed including <cpuid.h> on android (#10482) 2022-02-18 11:50:38 +03:00
Anton Voronov 61f915b4f6 [CPU] changed checks with_cpu_x86...() to mayiuse() (#9911) 2022-02-17 12:56:55 +03:00
Anton Voronov d5c837cc1b [CPU] added some legacy parallel methods to fix perf issues (#9758) 2022-02-10 15:13:05 +03:00
Anton Voronov b8fb666dbc [CPU] fixed dw conv fusing (#8193) 2021-10-29 11:09:51 +03:00
Anton Voronov 8280a1a045 [CPU] cleanup after oneDNN v2.3 migration (#7933) 2021-10-20 13:23:25 +03:00
Anton Voronov 4ef700c6e4 [CPU] fixed convolution outputShape in ConvDWConv fusing (#7239) 2021-08-26 13:38:24 +03:00
Anton Voronov ae25f5f581 [CPU] gemm inner product - memory access fix (#6072) 2021-07-14 16:09:41 +03:00
Anton Voronov dcce22bd81 [CPU] fixed conv + dw conv fusing (#6333) 2021-06-29 09:54:29 +03:00
Anton Voronov 81be8c94b0 [CPU] Deconvolution int8 fusing support (#5683) 2021-06-04 16:46:13 +03:00
Anton Voronov 5e4c3c4804 [CPU] Deconvolution int8 support (#5565) 2021-05-21 13:54:56 +03:00
Anton Voronov 18fdc4b694 [CPU] fixed SIGSEGV when moving unaligned data in Quantize node (Issue 51255) (#5262) 2021-04-29 11:41:33 +03:00
Anton Voronov 617a1024e1 [CPU] fixed dw conv fusing (#5146) 2021-04-12 22:32:00 +03:00
Anton Voronov 4d112b1f90 [CPU] returned old behavior for fp32 avx2 1x1 conv with dw conv fusing (#4484) 2021-03-29 13:55:13 +03:00
Anton Voronov 627c5e6d0e [CPU] fixed MergePermuteAndReorder optimization (#3318) 2020-11-24 19:14:53 +03:00
Anton Voronov 6467a9f5b8 [CPU] added MergePermuteAndReorder optimization + added test (#2519) 2020-11-17 09:04:49 +03:00
Anton Voronov be065c0144 [CPU] fixed strides filling in blocking desc (#2353) 2020-11-10 11:25:12 +03:00
Anton Voronov 0e34b392ee [CPU] Supported depthwise 6d, 7d, ..., added test (#971) 2020-09-10 08:33:38 +03:00
Anton Voronov 64bcbe59c7 [CPU] gemm convolution: added post ops JIT implementation (#1817) 2020-09-09 09:17:57 +03:00
Anton Voronov 7c95e8f8ff [CPU] fix: supported Tile with more than 5 dims (#2062) 2020-09-07 15:39:19 +03:00
Anton Voronov f3c7c731e3 [MKLDNN] Fixed bias datatype in jit_uni_dw_conv kernel (#1131) 2020-07-03 21:57:17 +03:00
Anton Voronov 9b32414747 [CPU] Fixed PostOpsIntBlobMemory filling (#1133) 2020-06-30 12:01:29 +03:00
Anton Voronov 629ca3a5d8 [CPU] Gemm node: supported precisions U8 and I8 and added tests 2020-06-09 14:11:18 +03:00
Anton Voronov 70c2058b61 [CPU] supported ShuffleChannels and added tests (#636) 2020-06-05 14:10:55 +03:00