1. 04 8月, 2022 3 次提交
    • Z
      Phi generate_proposals_v2 (#44436) · 566c80ff
      zhiboniu 提交于
      * phi_generate_proposals_v2
      
      * remove old kernels
      
      * optest add eager_check
      
      * del lod
      
      * update
      
      * update
      
      * update test_detection with_lod
      
      * update nms_util
      
      * remove old nms_util.h
      566c80ff
    • X
      mv fold & unpool to phi (#44836) · e9994f2e
      xiaoting 提交于
      * fix conflicts
      
      * mv unused file
      
      * revert backward.h
      
      * revert lu_unpack kernel
      
      * rm .cu file
      
      * Update lu_unpack_kernel.cc
      
      * format phi yaml
      e9994f2e
    • S
      opt allreduce (#44843) · 1f9e2742
      sneaxiy 提交于
      1f9e2742
  2. 03 8月, 2022 14 次提交
  3. 02 8月, 2022 7 次提交
    • L
      fix namespace of GPUContext (#44822) · 65f38869
      Leo Chen 提交于
      65f38869
    • S
      move renorm op (#44676) · 669353c1
      seemingwang 提交于
      * move renorm op
      
      * change python api
      
      * change op class func
      
      * alloc data
      
      * remove comments
      
      * fix grad arguments
      
      * fix python argument
      
      * fix python argument
      
      * change unit-test
      
      * remove shape func registration
      
      * recover extra-arguments
      
      * recover shape functor
      669353c1
    • W
      Multihead matmul fp16 (#44792) · 0fd8ee63
      Wilber 提交于
      * multihead matmul add fp16
      
      * fix windows error
      
      * fix rocm error
      
      * fix rocm error
      0fd8ee63
    • Y
      [Phi] Move QR to Phi (#44742) · 2cf2e786
      Yulong Ao 提交于
      * [Phi] Move Qr to the Phi
      
      * [Phi] Regiter the cpu grad kernel for qr
      
      * [Phi] Share the cuda kernels to lstsq
      
      * [Phi] Remove some improper inlcude files
      
      * [Phi] Modify codes based on the reviews
      
      * [Phi] Remove unecessary files and add the cuda_only comment
      
      * [Phi] Remove the unecessary include file
      
      * [Phi] Remove qr_op.cu and lstsq_op.cu
      2cf2e786
    • H
      [XPU] fp16 for layer_norm op (#44778) · 4c3e13de
      houj04 提交于
      * [XPU] fp16 for layer_norm op. test=kunlun
      4c3e13de
    • C
      [phi] add yolov3_loss yaml and unittest (#44476) · c7cf12fc
      ccrrong 提交于
      * add yaml and unittest
      
      * update yaml
      
      * update backward yaml and unittest
      
      * update yaml
      
      * add Yolov3LossGradInferMeta
      
      * update yolov3_loss_op.cc
      
      * fix bug
      
      * code format
      c7cf12fc
    • M
      support beam_search operator on xpu. test=kunlun (#44720) · 9bf80772
      mengqingchun02 提交于
      * support beam_search operator on xpu. test=kunlun
      
      * support beam_search operator on xpu. test=kunlun
      
      * support beam_search operator on xpu. test=kunlun
      
      * support beam_search operator on xpu. test=kunlun
      
      * support beam_search operator on xpu. test=kunlun
      9bf80772
  4. 01 8月, 2022 7 次提交
  5. 30 7月, 2022 1 次提交
  6. 29 7月, 2022 8 次提交