Skip to content

P
Paddle

机器未来 / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
切换分支/标签
  1. 19 12月, 2019 1 次提交
  3. 18 12月, 2019 2 次提交
  5. 17 12月, 2019 1 次提交
  7. 16 12月, 2019 6 次提交
  9. 15 12月, 2019 2 次提交
  11. 12 12月, 2019 4 次提交
  13. 11 12月, 2019 5 次提交
  15. 10 12月, 2019 10 次提交
  17. 09 12月, 2019 3 次提交
    • L
      QAT Int8 document (#21360) · fbf9eca0
      由 lidanqing 提交于 
      * update benchmark for int8v2, QAT1, QAT2 accuracy and performance
test=document_fix

* change according to reviews
test=develop test=document_fix

* improve some descriptions and some models
test=develop test=document_fix

* update models benchmark data
test=develop test=document_fix

* update int8v2 and qat2 performance
test=develop test=document_fix
      fbf9eca0
    • L
      Refine VarBase init function (#21587) · 4f81d1bd
      由 Leo Chen 提交于 
      * refine init function, test=develop

* add tests, test=develop

* remove extern, which may cause symbol error in gcc-4.8, test=develop
      4f81d1bd
    • L
      dygraph_grad_maker supports varbase without grad_var (#21524) · 84b72671
      由 Leo Chen 提交于 
      * dygraph_grad_maker supports varbase without grad_var, test=develop

* fix compile, test=develop

* fix test_tracer, test=develop

* follow comments, test=develop
      84b72671
  19. 07 12月, 2019 1 次提交
  21. 06 12月, 2019 5 次提交
    • Z
      Polish op registry codes (#21561) · 0f888836
      由 Zeng Jinle 提交于 
      * polish infer shape registry, test=develop

* modify some operators registry, test=develop
      0f888836
    • A
      Set lod_level of Out in compile time of sequence_pool_op (#21604) · 3d9dee57
      由 Aurelius84 提交于
      3d9dee57
    • Z
      refine dev_ctx.Wait() exception throw, test=develop (#21600) · 97e76cb9
      由 Zeng Jinle 提交于
      97e76cb9
    • H
      Add Much Complex Test and Fix Bugs for Control Flow cond API (#21532) · 1dcf6a72
      由 Huihuang Zheng 提交于 
      Add tests to use dy/dx to make sure the gradient values calculated by the control flow backward is correct. Also fixed bugs detected by those tests.

Fix bugs:

1. Unlike sum_op, optimizer ops don't allow uninitialized input tensor. But in conditional_block_grad_op, since the conditional_block may not run, the output gradient tensor may be uninitialized, which will cause the optimizer op error. To fix it, we should let optimizer ops support uninitialized input like sum_op or assign the uninitialized gradient to 0 when the conditional_block_grad_op doesn't run. I found there are about 10+ optimizer ops. **To be simpler, I just assign output gradient of the conditional_block_grad_op to 0 in this PR**. But it can be further explored whether we can make optimizer ops like sum_op to support uninitialized input tensor because theoretically we can speed up without the assigning in conditional_block_grad_op.

2. Infer parameter shapes during append_backward. I didn't know that all our parameters are in global block. When op_desc is inferring shapes at the sub-block, it may not know the shape of gradients of parameters whose shape information is at global block. I fixed it by inferring shapes of gradients from forward var.

This PR also did some code clean up:
1. Print the var name when sgd_op catches shape error so that it is easier to debug
2. Fix a typo: dicta -> dict
      1dcf6a72
    • H
      Paddlebox Related to Framework (#21586) · c5aec2fe
      由 hutuxian 提交于 
      * Add a single_process_multi_thread transpiler.
* Add some UTs.
* Fix some API description.
      c5aec2fe