1. 05 8月, 2022 1 次提交
  2. 04 8月, 2022 2 次提交
    • S
      Matmuls with activation and elementwise_add fuses (#44655) · 0420d514
      Sławomir Siwek 提交于
      * Add unit tests
      
      * matmul_v2 + activation
      
      * matmuls + elementwise_add
      
      * matmul_v2 postops
      
      * transform matmul to v2
      
      * opcompat
      
      * fix fusing matmul with multipe outs
      
      * add shape constraints
      
      * remove unused vars
      
      * change pass order
      
      * - Unit tests to be debugged
      
      - fix
      
      - refactor
      
      - diagnostic
      
      - more diagnostic
      
      - fix
      
      - Fix number two
      
      - fix
      
      - fix
      
      - fix
      
      - alpha added
      
      - more fixes
      
      - compilation fix
      
      - removed diagnostic code
      
      - cosmetic fixes
      
      * lint
      
      * add alpha constraint
      
      * merge matmul refactor
      
      * trigger CI
      
      * - fix
      
      * - another fix
      
      * code style
      
      * add support for matmul+elementwise_add+activation
      
      * code style
      
      * fix bfloat16 bugs
      
      * change append_binary to append_sum
      Co-authored-by: NJacek Czaja <jacek.czaja@intel.com>
      0420d514
    • J
      added conv and conv_tranpose support for md (#44677) · b2727020
      jakpiase 提交于
      b2727020
  3. 01 8月, 2022 1 次提交
  4. 29 7月, 2022 1 次提交
    • J
      [WIP] Matmul v1 & v2 unification -- part 1 (#44640) · 653885a5
      Jacek Czaja 提交于
      * - Unit tests to be debugged
      
      - fix
      
      - refactor
      
      - diagnostic
      
      - more diagnostic
      
      - fix
      
      - Fix number two
      
      - fix
      
      - fix
      
      - fix
      
      - alpha added
      
      - more fixes
      
      - compilation fix
      
      - removed diagnostic code
      
      - cosmetic fixes
      
      * lint
      653885a5
  5. 19 7月, 2022 1 次提交
    • J
      Added pad3d and pad2d FP32 FWD oneDNN kernels (#43990) · 2792b8de
      jakpiase 提交于
      * Piotrek's changes for pad3d
      
      * my changes
      
      * first version of pad3d, single copy, unnecessary reads
      
      * optimized pad3d kernel
      
      * test upadte
      
      * removed magic numbers
      
      * added support for pad2d
      
      * reverted two files
      
      * reverted one old change
      
      * added support for Paddings tensor
      
      * CI fix
      
      * CI fix
      
      * fixed timeout of tests
      
      * fixed typo
      
      * changes to GetKernelTypeForVar
      
      * Revert "changes to GetKernelTypeForVar"
      
      This reverts commit 469106115c49682b25038a666fd71bd4a10fb66b.
      
      * added AsExtra() to pad2d
      Co-authored-by: NPiotr Paturej <piotr.paturej@intel.com>
      2792b8de
  6. 14 7月, 2022 1 次提交
    • Y
      [Phi]Improve the mechanism for mkldnn kernel in PHI (#43941) · e9b4d0be
      YuanRisheng 提交于
      * adapt mkldnn kernel in PHI
      
      * fix ci compile bugs
      
      * fix compile bugs
      
      * fix compile bugs
      
      * fix compile bugs
      
      * fix compile bugs
      
      * delete comment
      
      * fix compile bugs in windows-inference
      
      * delete code for converage
      
      * modify code by review
      
      * modify code by review
      
      * add todo
      
      * fix compile bugs
      
      * fix compile bugs
      
      * fix compile bugs
      
      * fix unittest bugsx
      e9b4d0be
  7. 12 7月, 2022 1 次提交
    • S
      matmul+activation fuse pass (#43519) · 3333a439
      Sławomir Siwek 提交于
      * add method for post ops
      
      * format code
      
      * gpd
      
      * format style
      
      * add matmul+act test
      
      * implement matmul+activation
      
      * whitespaces
      
      * code style
      
      * python code format
      
      * Increase UT timeout
      
      * code format
      
      * update style
      
      * generalize activation fuse passes
      
      * change order
      
      * Unify activation GPD
      
      * Revert changes with op_act
      
      * remove softmax mkldnn attrs
      
      * set common name for act attributes
      
      * whitespace
      
      * append postops by helper function
      
      * ut style
      
      * revert changes related to quantization
      
      * Reduce redundancy
      
      * reduce number of parameters
      
      * trigger CI
      
      * validate attribute
      
      * trim unit test
      3333a439
  8. 11 7月, 2022 1 次提交
  9. 06 7月, 2022 2 次提交
    • J
      Performance fix for recommender model (#43803) · 48abaec6
      jakpiase 提交于
      * fix for binary kernels
      
      * fixed performance for elementwise, reduce and concat
      
      * added comment
      
      * CI fix
      
      * CI fix
      
      * added formatting
      
      * reverted one file
      
      * Revert "reverted one file"
      
      This reverts commit 54725e1c62318d3a18913821200e973816751019.
      
      * Revert "added formatting"
      
      This reverts commit b9795dd253d755a329376d7ab0542860aa7815c6.
      
      * added enforcing oneDNN BF16 reduce kernel
      
      * fix for eltwise and reenabled reshape kernels
      
      * fix for binary handler
      
      * added formatting
      
      * referted changes for flatten,squeeze and reshape ops
      48abaec6
    • J
      Set FC input data format to ANY (#44023) · aa18ae11
      joanna.wozna.intel 提交于
      * Fc add any to input format
      
      * Pre-commit changes
      aa18ae11
  10. 05 7月, 2022 1 次提交
  11. 04 7月, 2022 1 次提交
  12. 02 7月, 2022 2 次提交
    • L
      unify cpu context, part2 (#44012) · 755438a7
      Leo Chen 提交于
      * fix init()
      
      * delete test_device_context
      
      * replace CPUDeviceContext with CPUContext
      
      * fix test_scalar
      
      * remove dot_op.cc
      
      * fix compile
      755438a7
    • L
      unify cpu context (#43989) · 09096aeb
      Leo Chen 提交于
      * unify cpu context
      
      * fix init()
      
      * delete test_device_context
      
      * fix test_scalar
      09096aeb
  13. 26 6月, 2022 1 次提交
  14. 23 6月, 2022 2 次提交
    • W
      【code format】Fix cpplint style 4 (#43695) · 5a55f13b
      wangzhen38 提交于
      * cpplint fix 2
      
      * cpplint fix 2
      
      * fix cpplint style 4
      
      * fix cpplint style 4
      
      * fix cpplint style 4
      
      * fix cpplint style 4
      5a55f13b
    • S
      [external reviewing] Params to int8 pass (#42625) · b8b2d6a9
      Sylwester Fraczek 提交于
      * sylwek
      
      prototype params to int8 pass
      
      * trying to make warmup work
      
      * wip
      
      * wip
      
      * change test to cpp test
      
      * review fixes, refactoring
      
      * more refactoring
      
      * add erasevars
      
      * change test to fixture
      
      * rename pass
      
      and reorder erasevars and graphsaferemovenodes
      
      * fix
      
      * more refactoring and fixed bug
      
      * formatting
      
      * remove scale count
      
      * enfroce message too short
      
      * remove erasevars
      
      erasevars couldbe cauuse of memory issues
      
      some other fixes
      
      * add count of successfull fuses to name of new nodes
      
      * FindVar -> GetVar and use ConvResidual pattern
      
      * use tensor->clear() instead of new variable
      
      * Update paddle/fluid/framework/ir/mkldnn/params_quantization_mkldnn_pass_tester.cc
      Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>
      
      * Update paddle/fluid/framework/ir/mkldnn/params_quantization_mkldnn_pass_tester.cc
      Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>
      
      * Update paddle/fluid/inference/tests/api/analyzer_lexical_analysis_gru_tester.cc
      Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>
      
      * add log (review fix)c
      
      * review fix (2 functions to one)
      
      * code review: Conv->QuantizeConv
      
      * revert
      
      * fix formatting
      
      * remove unused functions
      
      * add paddle enforce
      Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>
      b8b2d6a9
  15. 21 6月, 2022 1 次提交
    • S
      Generalize conv+activation fuse pass (#43382) · 347e4b2e
      Sławomir Siwek 提交于
      * consolidate conv act passes
      
      * generalize conv_activation
      
      * integrate conv+act tests
      
      * code style format
      
      * whitespaces
      
      * remove timeout from old tests
      
      * implement comments from review
      
      * restore ut
      
      * whitespace
      
      * code style
      
      * transpose
      
      * fixes after review
      
      * method for gettin act
      
      * Change Paddle_enforce error type
      
      * code format
      
      * add missing opcompats
      347e4b2e
  16. 20 6月, 2022 1 次提交
  17. 16 6月, 2022 2 次提交
  18. 07 6月, 2022 1 次提交
  19. 05 6月, 2022 1 次提交
  20. 04 6月, 2022 1 次提交
  21. 31 5月, 2022 1 次提交
  22. 25 5月, 2022 1 次提交
  23. 23 5月, 2022 1 次提交
    • J
      [Internal reviewing] NHWC fix to am_vocoder model for oneDNN 2.6 (#42729) · d414af94
      Jacek Czaja 提交于
      * - prototype of reimplemented fixes
      
      * - compilation fixes
      
      * - compilation fix
      
      * - cosmetic info
      
      * - hopefully fix
      
      * - compilation fix
      
      * - supported for nested blocking of cache clearing
      
      * - fix
      
      * - Unit test to changes
      
      * - Compilation fix to windows (hopefully)
      
      * - Moved resetting layout to ResetBlob
      
      * - fixes after review
      d414af94
  24. 19 5月, 2022 1 次提交
  25. 18 5月, 2022 1 次提交
  26. 11 5月, 2022 1 次提交
    • Z
      Move weights and biases scale computing into pass (#42241) · c0652972
      Zuza Gawrysiak 提交于
      * Add int8 scales gathering pass for convolution
      
      * Fix typo
      
      * Add unittest
      
      * Add corrected unit test
      
      * Change test name
      
      * Remove enabling mkldnn in test
      
      * Speed up test
      
      * Change max examples
      
      * Add functional test
      
      * Change test name
      
      * Add new test case
      
      * Rename pass
      c0652972
  27. 10 5月, 2022 2 次提交
  28. 05 5月, 2022 1 次提交
  29. 21 4月, 2022 1 次提交
  30. 19 4月, 2022 1 次提交
  31. 14 4月, 2022 2 次提交
    • S
      FC+elementwise_add (residual connection) (#41776) · 92d8d0bc
      Sławomir Siwek 提交于
      * Change tensor name to match activation
      
      * declare fc_eltwise_add pass
      
      * merge conv_eltwise refactor PR
      
      * first compilable draft
      
      * unittest feedback tools
      
      * Fuse pass tester
      
      * Move IsReachable() to shared file
      
      * 100% coverage of fuse_pass_tester.cc
      
      * register pass
      
      * Add bias node
      
      * Improve unit tests / remove bias node from pattern
      
      * improve fc_eltwiseadd_unittest
      
      * cancel eltwise_add fuse if act is already fused
      
      * Add elementwise_input scale
      
      * Residual MVP
      
      * Add new FC attrs
      
      * Add more test cases
      
      * Add missing op attrs
      
      * Adapt code to new Elementwise pattern
      
      * reuse existing fcpattern
      
      * improve code style
      
      * remove unused arguments
      
      * fix typo
      
      * remove whitespace
      
      * remove int8 related code
      
      * Remove attributes from base ops
      
      * style
      
      * style check
      
      * Remove input from base op
      
      * Set attribute during fuse
      
      * ut timeout
      
      * download and test model
      
      * DRY
      
      * apply feedback from review
      
      * Style check
      
      * fix typo
      
      * cosmetic changes
      
      * explicitly set residual as output
      
      * VIT-OCR accuracy check
      
      * trigger CI
      
      * remove whitespaces
      
      * fix missing data file
      92d8d0bc
    • J
      Added shuffle_channel BF16/FP32 FWD oneDNN kernel (#39756) · c7623d72
      jakpiase 提交于
      * added shuffle_channel bf16/fp32 fwd kernel
      
      * added missing files
      
      * CI fix
      
      * changed from pten to phi
      
      * tmp save
      
      * added reviewers suggestions
      
      * fix for test
      c7623d72
  32. 11 4月, 2022 1 次提交
  33. 30 3月, 2022 1 次提交