Skip to content

P
Paddle

Crayon鑫 / 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. 16 4月, 2019 2 次提交
  3. 15 4月, 2019 3 次提交
  5. 10 4月, 2019 2 次提交
  7. 09 4月, 2019 3 次提交
  9. 08 4月, 2019 3 次提交
    • Q
      Imperative fix bugs (cherry pick to 1.4) (#16680) · 7e560558
      由 Qiyang Min 提交于 
      * Fix imperative bugs

test=release/1.4

* Polish code

test=release/1.4
      7e560558
    • J
      Cherry pick/fix transformer (#16620) · 29f34416
      由 Jiabin Yang 提交于 
      * Imperative deep-first backward process (#16605)

* Fix bug of gradient interface

* shrink transformer

* Right transformer

* Change from width-first backward to deep-first backward process

test=develop

* Reverse iterator op's input

test=develop

* Polish code

* Change the iteration direction in ingrads' map slots

test=develop

* Polish code

test=develop

* test=develop, cherry-pick fix for transformer in dygraph

* test=develop, fix transformer in dygraph
/
      29f34416
    • J
      Cherry pick 1.4/ptb fix (#16607) · 89d09b83
      由 Jiabin Yang 提交于 
      * test=develop, ptb_rnn fix op

* test=release/1.4, refine code

* test=release/1.4, fix ci failed error
      89d09b83
  11. 07 4月, 2019 1 次提交
  13. 04 4月, 2019 1 次提交
  15. 03 4月, 2019 1 次提交
  17. 31 3月, 2019 2 次提交
    • Q
      Add linear learning warmup method in learning rate scheduler. (#16563) · 1ebd7434
      由 qingqing01 提交于 
      * Add linear learning warmup method

This warmup lr can be combinated with other learning rate strategies.
For example:
            decayed_lr = fluid.layers.linear_lr_warmup(
                fluid.layers.piecewise_decay(boundaries, lr_steps),
                warmup_steps, start_lr, end_lr)
      1ebd7434
    • W
      Batch norm cudnn accurate (#16545) · 22b02bfa
      由 Wu Yi 提交于 
      * fix cudnn batch norm accuracy test=develop

* fix cudnn batch norm accuracy test=develop

* disable failed test for later fix test=develop
      22b02bfa
  19. 30 3月, 2019 3 次提交
  21. 29 3月, 2019 19 次提交