1. 31 5月, 2022 2 次提交
    • X
      [EinsumOp] Make EinsumOp support bfloat16. (#43085) · a4bb38cb
      xiongkun 提交于
      * change einsum_v2 as default and add new flags: FLAG_einsum_opt=1|0
      
      * make EInsumOP support bf16
      
      * add unittest for BF16
      
      * add condition for test_BF16
      
      * fix bugs
      
      * fix
      a4bb38cb
    • Z
      add embedding yaml (#43029) · 2785f876
      zyfncg 提交于
      * add embedding yaml
      
      * fix infermeta bug
      
      * fix bug of selected_rows infer_meta
      
      * fix selected_rows
      
      * add unittest
      2785f876
  2. 30 5月, 2022 5 次提交
  3. 27 5月, 2022 2 次提交
  4. 26 5月, 2022 2 次提交
  5. 25 5月, 2022 2 次提交
    • L
      fix maybe-uninitialized warning (#42902) · f1f79b0d
      Leo Chen 提交于
      * fix maybe-uninitialized warning
      
      * fix compile
      
      * fix xpu compile
      
      * fix npu compile
      
      * fix infer compile
      
      * fix compile
      
      * fix compile
      f1f79b0d
    • X
      [EinsumOp] Optimize the backward speed of EinsumOp (#42663) · 71b046cd
      xiongkun 提交于
      * change logic for optimize
      
      * modifty
      
      * optimize the backward speed of EinsumOp
      
      * add cache optimizer for einsum op
      
      * EinsumOp: fix new dygraph mode error
      
      * fix bug
      
      * change Cache->InnerCache
      
      * fix code
      
      * fix
      
      * add nan inf utils for einsum op
      
      * add as_extra
      
      * Compatible with v2.3 EinsumOp
      
      * remove dispensable
      71b046cd
  6. 24 5月, 2022 2 次提交
  7. 23 5月, 2022 4 次提交
  8. 20 5月, 2022 5 次提交
    • N
      Delete ElementwiseKernel in BroadcastKernel (#42779) · 0d878f1a
      niuliling123 提交于
      0d878f1a
    • L
      use fp32 compute type for cublasGemmStridedBatchedEx with fp16 input/output (#42851) · f36a9464
      Leo Chen 提交于
      * use fp32 compute type for cublasGemmStridedBatchedEx with fp16 input/output
      
      * add flags to control compute type
      
      * default to false
      
      * add unit test
      
      * default to true
      f36a9464
    • Y
      move activation kernel (#42880) · 191c441a
      YuanRisheng 提交于
      191c441a
    • W
      [Eager] Make CreateInferMeta more robust (#42871) · d8b69124
      Weilong Wu 提交于
      d8b69124
    • X
      [Hackathon No.5] tril_indices OP (#41639) · 75db5b86
      xiaoguoguo626807 提交于
      * add tril_indices cpu kernal
      
      * modify tril_indice cpu op
      
      * modify bug
      
      * modify bug
      
      * add tril_indices python api
      
      * add tril_indices python api
      
      * resolve conflict
      
      * add tril_indices test
      
      * modify details
      
      * add tril_indices.cu
      
      * pythonapi pass
      
      * save tril_indices
      
      * CPU tril_indices pass
      
      * delete vlog
      
      * modify test_tril_indices_op.py
      
      * delete tril_indices_kernel.cc.swp
      
      * delete tril_indice.cu
      
      * modify code style
      
      * add newline in creation.py
      
      * modify creation.py linux newline
      
      * delete annotation
      
      * check code style
      
      * check .py style add final_state??
      
      * modify code style
      
      * add gpu_tril_indices
      
      * modify gpu_compiled_juage
      
      * modify gpu judge
      
      * code style
      
      * add test example
      
      * modify english document
      
      modify english document
      
      modify english document
      
      modify document
      
      modify document
      
      * modify pram name
      
      * modify pram name
      
      * modify pram
      
      * reduce test ex
      75db5b86
  9. 19 5月, 2022 3 次提交
  10. 18 5月, 2022 4 次提交
  11. 17 5月, 2022 1 次提交
  12. 16 5月, 2022 3 次提交
  13. 13 5月, 2022 1 次提交
  14. 12 5月, 2022 2 次提交
  15. 11 5月, 2022 1 次提交
    • Z
      [Phi] Change the output format of C++ backward api (Part1) (#42677) · ba71fbea
      zyfncg 提交于
      * change the output format of C++ backward api
      
      * fix merge conflict
      
      * fix sparse api code auto-gen
      
      * fix eager_gen bug
      
      * fix bug of output is null
      
      * fix bug of conv2d_grad_impl
      
      * fix optional grad
      
      * fix bug of eager-gen double_grad
      
      * fix bug
      
      * fix multiply_double_grad bug
      
      * remove node pruning
      ba71fbea
  16. 10 5月, 2022 1 次提交