Skip to content

P
Paddle

PaddlePaddle / Paddle
1 年多 前同步成功

通知 2302
Star 20931
Fork 5422
P
Paddle
切换分支/标签
  1. 14 11月, 2017 2 次提交
  3. 23 10月, 2017 1 次提交
  5. 17 10月, 2017 1 次提交
  7. 10 10月, 2017 1 次提交
  9. 17 9月, 2017 1 次提交
  11. 15 9月, 2017 2 次提交
  13. 14 9月, 2017 1 次提交
  15. 12 9月, 2017 1 次提交
  17. 30 8月, 2017 1 次提交
  19. 23 8月, 2017 1 次提交
  21. 08 8月, 2017 2 次提交
  23. 24 7月, 2017 1 次提交
  25. 21 7月, 2017 3 次提交
  27. 17 7月, 2017 1 次提交
  29. 14 7月, 2017 1 次提交
  31. 12 7月, 2017 1 次提交
  33. 11 7月, 2017 1 次提交
    • Y
      Fix slow parsing a recursive depends topology · 313e9f55
      由 Yu Yang 提交于 
      * Fix  #2797
* It because trainer_config_helpers' __dfs_travel__ did not record the
  node which travelled, and if the topology has a recursive dependency,
  there are some nodes will be travelled multiple times.
* Add a `travelled` set to record which node is travelled.
* Also add a unittest for this situation.
      313e9f55
  35. 10 7月, 2017 1 次提交
  37. 03 7月, 2017 1 次提交
  39. 29 6月, 2017 3 次提交
  41. 27 6月, 2017 1 次提交
  43. 24 6月, 2017 1 次提交
  45. 22 6月, 2017 2 次提交
  47. 07 6月, 2017 1 次提交
  49. 26 4月, 2017 1 次提交
    • Y
      Add error clipping to MT demo. · 3f1151a5
      由 Yu Yang 提交于 
      * Compose GRU step naive layer in trainer config helpers.
  * It is uses mixed_layer for gate.
  * It supports ERROR_CLIPPING, DROPOUT
* Add error clipping in MT demo.
* Fix #1143
* Fix #1891
      3f1151a5
  51. 02 2月, 2017 1 次提交
  53. 31 1月, 2017 2 次提交
  55. 09 12月, 2016 1 次提交
  57. 12 11月, 2016 1 次提交
  59. 02 11月, 2016 1 次提交
    • Q
      Add job=time in trainer, refine cudnn_conv to reduce gpu memory and speed up training. (#218) · 45c81a41
      由 qingqing01 提交于 
      * Add benchmark for PaddlePaddle, tensorflow and caffe

* ConvProjection to reduce memory for goolenet

* Add unit test for ConvProjection.
1. unit test in test_LayerGrad.
2. compare the ConvPorjection and CudnnConvLayer, also compare the concat_layer+img_conv_layer and concat_layer_conv_projection.

* Reduce cudnn_conv memory and add benchmark document.
1. Use TmpMatrix as the workspace in cudnn_conv to reduce gpu memory. It reduce lots of memory.
2. Add benchmark document.
3. fix smallnet_mnist_cifar.py in paddle.

* Add job=time and refine cudnn_conv to reduce gpu memroy and speed up

* Refine cudnn_conv and shared biases operation in concat_layer and mixed_layer.

* follow comments

* follow comments

* Use unique_ptr to prevent memory leaks in CudnnConvLayer.
      45c81a41
  61. 24 10月, 2016 1 次提交