1. 05 12月, 2022 16 次提交
  2. 04 12月, 2022 2 次提交
  3. 03 12月, 2022 2 次提交
  4. 02 12月, 2022 20 次提交
    • C
      remove softmax api from fluid (#48388) · a3ae080a
      Charles-hit 提交于
      * move softmax to paddle2.0
      
      * fix some bugs
      
      * resolve conflict
      
      * remove some code
      
      * modify code style
      
      * fix bugs
      
      * fix code
      
      * fix move code
      
      * fix some bugs
      
      * fix code
      
      * fix some code
      
      * modify the header file
      
      * fix bugs
      
      * fix some examples
      
      * fix mish example
      
      * fix code
      a3ae080a
    • Y
    • R
      add custom profier doc (#48629) · d969c309
      ronnywang 提交于
      * add custom profier doc
      
      * update
      
      * format; test=document_fix
      Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>
      d969c309
    • P
      [PHI] Migrate elementwise_sub kernel (#48611) · 493825a5
      Piotr Paturej 提交于
      * Add migrations
      
      * Fix build errors
      
      * Remove elementwise_mul from migration
      493825a5
    • H
      Migrate mul_mkldnn_op to phi matmul_kernel (#48299) · e8edbb09
      Hulek 提交于
      * Migrate mul_mkldnn_op to matmul_kernel
      
      * Review fixes - changed mutable_data, changed ctx to dev_ctx, fixed namespaces
      
      * switched some funcs to phi
      
      * Deleted not needed phi:: and changed place checking according to standards
      e8edbb09
    • J
      [XPU ]Fix xpu compile error (#48621) · 2af82190
      Jiabin Yang 提交于
      * [Eager] Fix paddle.grad interface
      
      * [Eager] Support minimum SubGraph for GeneralGrad
      
      * Add needed_nodes to prune grad graph more thoroughly
      
      * [Eager] Add grad_node_trans_mapping_ to record which grad_node has been transformed to AccumulationNode
      
      * [Eager] Fix paddle.grad interface
      
      * Polish code
      
      * remove potential_stop_node
      
      * Add endding_nodes to enhance genSugraph logic
      
      * clear endding_nodes_
      
      * polish code
      
      * rename endding_nodes to endding_nades_
      
      * Refactor grad interface
      
      * Add register_hook case to fix coverage-ci
      
      * Fix code format
      
      * Refactor general_grad
      
      * Add more code comments
      
      * call clear directly to release GradSlotMeta
      
      * fix a mistake
      
      * fix matmul/ multiply kernel logic and optional input in yaml, fill zeros logic and so on.
      
      * fix batch_norm_double_grad yaml optional config
      
      * fix tanh_triple_grad yaml and kernels
      
      * fix MultiplyTripleGradKernel optional logic
      
      * fix merge mistake
      
      * fix compile error
      
      * remove legacy attr for bn
      
      * polish code
      
      * fix some kernel
      
      * merge develop
      
      * fix error
      
      * remote log
      
      * fix kernel with full like
      
      * hide value log behind
      
      * hide value log behind
      
      * fix matmul_triple grad
      
      * fix xpu compile error
      
      * fix xpu compile error
      
      * fix xpu ut
      
      * fix xpu ut
      
      * fix_xpu_compile_error
      Co-authored-by: NWeilong Wu <veyron_wu@163.com>
      2af82190
    • Z
      [fluid remove]: remove paddle.fluid.layers.nn.temporal_shift and... · 49656af3
      zqw_1997 提交于
      [fluid remove]: remove paddle.fluid.layers.nn.temporal_shift and paddle.fluid.layers.sigmoid_focal_loss (#48553)
      
      * remove paddle.fluid.layers.nn.temporal_shift
      
      * code check
      
      * rm unittest
      
      * remove paddle.fluid.layers.sigmoid_focal_loss
      49656af3
    • W
      [remove fluid] drop_out API (#48586) · 6af7b42b
      wangzhen38 提交于
      * [remove fluid] drop_out PI
      
      * [remove fluid] drop_out PI
      
      * [remove fluid] drop_out layernorm
      
      * [remove fluid] drop_out layernorm
      
      * [remove fluid] drop_out layernorm
      
      * [remove fluid] drop_out layernorm
      6af7b42b
    • H
      [Fluid Clean] remove argsort, linspace, diag (#48456) · 23299c70
      heyanru 提交于
      23299c70
    • B
      Split common funcs from reduction and structure modification (#46970) · ef575d6a
      Bo Zhang 提交于
      * profile reduce kernel for fp16 and reduceHigherdim
      
      * use reinterpret_cast
      
      * fix for CI on ROCm
      
      * add Macro for ROCm
      
      * ROCm CI config
      
      * ROCm CI config
      
      * unit test repair
      
      * pull
      
      * add common_funcs.h
      
      * reduceType
      
      * Update reduce_function.h
      
      * not higher
      
      * rename
      ef575d6a
    • L
      remove less_than (#48584) · e48767fe
      LiYuRio 提交于
      e48767fe
    • S
      Fix fuse_gemm_epilogue (#47805) · 6efc2888
      Shijie 提交于
      * Fix fuse_gemm_epilogue
      
      * update tests
      
      * Update CMakeLists.txt
      
      * Update CMakeLists.txt
      
      * Update CMakeLists.txt
      
      * fix random seed
      
      * use assert_allclose
      
      * Update test_dist_fuse_gemm_epilogue_pass.py
      
      * Update cpp_pass.py
      
      * Update test_dist_fuse_gemm_epilogue_pass.py
      
      * fix codestyle
      
      * update seed and atol
      6efc2888
    • G
      add some compare and logical trt converter (#48592) · 4c38b87e
      gem5 提交于
      4c38b87e
    • C
      fix unit test (#48624) · fcf26279
      Charles-hit 提交于
      fix test_primapi
      fcf26279
    • R
      fix phi capi kernel registration macro error (#48616) · 0f3b1ad6
      ronnywang 提交于
      * fix capi kernel registration macro error
      
      * update
      0f3b1ad6
    • V
      [Clean Fluid API]Remove API: pad (#48100) · a7c43ffa
      Vvsmile 提交于
      * replace pad with paddle.nn.functional.pad and fix call arguments
      
      * solve the ImportError in PR-CI-Mac-Python3, which is called "cannot
      import name 'assign' from 'paddle.fluid.layers'"
      
      * fix the CI error of remove_pad by remove import paddle.nn.functional as
      F
      
      * fix the ImportError of remove_pad
      
      * fix functional call of pad
      a7c43ffa
    • C
      move paddle.fluid.layers.tensor.create_parameter to... · 33d90ae8
      cyber-pioneer 提交于
      move paddle.fluid.layers.tensor.create_parameter to paddle.tensor.creation.create_parameter (#48579)
      
      33d90ae8
    • W
      [Eager, Performance Optimization] modify AllocateFrom to reduce deconstruction... · 708c4f88
      Weilong Wu 提交于
      [Eager, Performance Optimization] modify AllocateFrom to reduce deconstruction of shared_ptr (#48548)
      
      708c4f88
    • J
      [Eager] Optimize Grad by prune useless branch (#47827) · d1e93be1
      Jiabin Yang 提交于
      * [Eager] Fix paddle.grad interface
      
      * [Eager] Support minimum SubGraph for GeneralGrad
      
      * Add needed_nodes to prune grad graph more thoroughly
      
      * [Eager] Add grad_node_trans_mapping_ to record which grad_node has been transformed to AccumulationNode
      
      * [Eager] Fix paddle.grad interface
      
      * Polish code
      
      * remove potential_stop_node
      
      * Add endding_nodes to enhance genSugraph logic
      
      * clear endding_nodes_
      
      * polish code
      
      * rename endding_nodes to endding_nades_
      
      * Refactor grad interface
      
      * Add register_hook case to fix coverage-ci
      
      * Fix code format
      
      * Refactor general_grad
      
      * Add more code comments
      
      * call clear directly to release GradSlotMeta
      
      * fix a mistake
      
      * fix matmul/ multiply kernel logic and optional input in yaml, fill zeros logic and so on.
      
      * fix batch_norm_double_grad yaml optional config
      
      * fix tanh_triple_grad yaml and kernels
      
      * fix MultiplyTripleGradKernel optional logic
      
      * fix merge mistake
      
      * fix compile error
      
      * remove legacy attr for bn
      
      * polish code
      
      * fix some kernel
      
      * merge develop
      
      * fix error
      
      * remote log
      
      * fix kernel with full like
      
      * hide value log behind
      
      * hide value log behind
      
      * fix matmul_triple grad
      Co-authored-by: NWeilong Wu <veyron_wu@163.com>
      d1e93be1
    • H