1. 02 12月, 2020 15 次提交
    • C
      a2e9d95a
    • S
      enforce the matmul_v2 error message (#29297) · 0fb18bc2
      ShenLiang 提交于
      0fb18bc2
    • Z
      Remove some useless log. (#29300) · 9b59a589
      Zhen Wang 提交于
      9b59a589
    • L
      fix shape of tile_grad op (#29289) · 13a22a37
      Leo Chen 提交于
      13a22a37
    • Z
      Add pure fp16 training with master weights. (#27712) · be3777a5
      Zhen Wang 提交于
      * add the weight decay func for the momentum op
      
      * Add the multi_precision function in Momentum Optimizer.
      
      * Make sure that the initial value of master weights are same with the fp16 weights.
      
      * add static loss scaling.
      
      * add the rescale_grad function in the pure fp16 training.
      
      * use the original momentum updating method.
      
      * Polish some codes, such as variable names.
      
      * add docstring for apis.
      
      * update the var creation details of _create_master_weight.
      
      * not modify codes about imperative momentum updating.
      
      * Fix the error of test_dist_sparse_tensor_load_momentum UT.
      
      * add unit test for multi precision fp16 training.
      
      * add more unit tests for CI.
      
      * Use lower threshold values for allclose comparing in test_multi_precision_fp16_train UT.
      
      * For CI Coverage Checking.
      be3777a5
    • C
      fix random failed of complex matmul (#29285) · 976961de
      chentianyu03 提交于
      976961de
    • W
      change import math.h to cmath (#29260) · 6673fb05
      Wojciech Uss 提交于
      6673fb05
    • F
      Layer norm fp16 (#29169) · 7584bb50
      furnace 提交于
      * add fp16 for layer_norm op
      
      * revert layernorm api
      
      * fix forward
      
      * fix forward
      
      * fix backward for layernorm with fp16
      
      * fix unit test for layernorm with fp16
      
      * fix with_mkldnn compile error for layernorm with fp16
      
      * 1. revert to PADDLE_ENFORCE_NOT_NULL, 2. change static_cast<float> to static_cast<U>
      
      * fix with_mkldnn compile error for layernorm with fp16
      
      * fix with_mkldnn compile error for layernorm with fp16
      Co-authored-by: Nzhiqiu <chenqiuliang@baidu.com>
      7584bb50
    • C
      Supprot precision test for code analysis · 597897e3
      chalsliu 提交于
      597897e3
    • M
      Update APIs in text/datasets and dataloader (#29219) · a37963b8
      mls1999725 提交于
      * Update IterableDataset API
      
      * Update TensorDataset API
      
      * Update APIs in paddle/text/datasets
      
      * Update dataset.py
      a37963b8
    • M
      Update Codes of Cifar and VOC2012 (#29204) · 493568b0
      mls1999725 提交于
      * Update Cifar Codes
      
      * Update VOC2012 Codes
      
      * Update voc2012.py
      
      * Update voc2012.py
      
      * Update cifar.py
      
      * Update cifar.py
      
      * Update voc2012.py
      493568b0
    • M
      Update get_worker_info API (#29190) · 0aedd463
      mls1999725 提交于
      * Update get_worker_info API
      
      * Update dataloader_iter.py
      
      * Update dataloader_iter.py
      
      * Update dataloader_iter.py
      0aedd463
    • S
    • M
      Update conv3d API (#29205) · 6a9a62c3
      mls1999725 提交于
      * Update conv3d API
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      
      * Update nn.py
      6a9a62c3
    • H
      [Dy2stat] Fix PaddleGan Deoldify Model Dy2stat Problems (#29226) · aec05d81
      Huihuang Zheng 提交于
      This PR fixes several problems in dy2stat for Deoldify model in PaddleGan.
      
      In model, software engineer wrote if x.shape == y.shape, the Tenser shape is a tuple in dygraph so the == returns True/False, but in static graph the == becomes element-wise comparison, which is a different behavior. In this PR we reduce the element-wise comparison result.
      
      If software engineer write computations which uses parameters in hooks, the static graph can loss the parameter variable because we put param_guard at forward of a Layer. In this PR we made param_guard cover pre-hook and post-hook.
      
      In PaddleGan, software engineer calculated some parameter values in __init__ by running some dygraph code. Those code also run during dy2stat. So some variables may be assign as a VarBase (Tensor) first and then Variable, which raised an error. We fixed the bug in this PR by handling the case.
      
      TODO: We just added testcase for the 1. shape comparison. Should add test case for 2. and 3. But since we are chasing 2.0RC, I will do it in the near future PR
      aec05d81
  2. 01 12月, 2020 20 次提交
  3. 30 11月, 2020 5 次提交