1. 15 4月, 2022 14 次提交
    • F
      [MLU] add mlu softmax kernel (#41816) · 2d6b71a2
      fwenguang 提交于
      2d6b71a2
    • J
      Add eager string tensor (#41039) · a22b68b8
      Jack Zhou 提交于
      * Add core.eager.StringTensor __init__ which pyarray args can be passed
      
      * Add the numpy method of core.eager.StringTensor
      
      * revert tensor.to_string modification
      
      * Add ToPyObject for core.eager.StringTensor
      
      * Add debug string for core.eager.StringTensor
      
      * Remove place args of core.eager.StringTensor temporarily
      
      * Fix check string_tensor error
      
      * remove dtype of core.eager.StringTensor
      
      * add core.eager.StringTensor unittest
      
      * remove pstring from VarDesc
      
      * Add InitStringTensorWithStringTensor
      
      * Remove to_string modification
      
      * Remove zero_copy arg from StringTensor creator
      a22b68b8
    • Z
      [XPUPS]fix hashtable_kernel.kps (#41790) · ef6ff4ef
      zmxdream 提交于
      * refactor heter comm kernel
      
      * update. test=develop
      
      * update calc_shard_offset. test=develop
      
      * update xpu kernel. test=develop
      
      * update args of calc_shard_offset
      
      * update. test=develop
      
      * remove customGradMerger
      
      * update. test=develop
      
      * update. test=develop
      
      * fix. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update optimizer kernel
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * add optimizer kernel. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix kunlun not support size_t. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * update hashtable. test=develop
      
      * update. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * update. test=develop
      
      * update. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * template init. test=develop
      
      * hashtable template init. test=develop
      
      * fix. test=develop
      
      * fix. test=devlop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix hashtable_kernel. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      
      * fix. test=develop
      Co-authored-by: NWorgenZhang <frank08081993@gmail.com>
      ef6ff4ef
    • A
      [IPU] add mixed-precission support for ipu (#41733) · d7224482
      Allen Guo 提交于
      * add mixed-precission support for ipu
      
      * restore cast_model_to_fp16 api
      
      * update UTs
      d7224482
    • C
      polish tensor depreacted method warning (#41807) · e83e44c7
      Chen Weihang 提交于
      e83e44c7
    • Z
      Add API: Sparse Convolution3D (#41434) · 1665594d
      zhangkaihuo 提交于
      1665594d
    • P
      support no_need_buffer in eager_fluid state (#41720) · 840d2eb6
      pangyoki 提交于
      * support no_need_buffer in eager_fluid state
      
      * change no_need_buffer info from fwd_info to bwd_info
      
      * fix CI fail, gru_unit donnot use no_need_buffer
      
      * fix conflict between no_need_buffer and dispensable
      
      * use tensor.define in dispensable
      
      * solve conflict
      
      * solve conflict
      840d2eb6
    • A
    • Z
    • L
      Change cuDNN Conv kernel for auto tune feature (#41313) · 35acfeda
      limingshu 提交于
      * change cudnn helper for auto-tune
      
      * Add FLAGS_use_autotune to set the global status of autotune and change the order of choosing algorithm.
      
      * Fix the bug in calculating and printing current step cache hit rate.
      
      * Improve the autotune cache and fix unittest.
      
      * Change the key from AlgorithmType to int64_t.
      
      * Fix unittest for cpu-only env.
      
      * change ChooseAlgoByWorkspace for heuristic mode
      Co-authored-by: NLiu Yiqun <liuyiqun01@baidu.com>
      35acfeda
    • F
      [MLU] add mlu activation kernels (#41751) · 10114859
      fwenguang 提交于
      10114859
    • F
      [MLU] add mlu new profiler (#41138) · fc208b7e
      fwenguang 提交于
      * [MLU] add mlu new profiler
      
      * fix format
      fc208b7e
    • C
      [Auto Parallel]update cluster (#41722) · 605552a9
      caozhou 提交于
      * update cluster
      605552a9
    • H
      fix batch norm memory issue (#41717) · 42abcc08
      hong 提交于
      * try to fix batch norm memory issue
      
      * fix batch norm memroy alloc bug
      
      * polish some code
      42abcc08
  2. 14 4月, 2022 26 次提交