RhysWu issues

Repositories
Issues
Comments

Results 2 issues of


                                            RhysWu

Do I have to rewrite test.py when I define the network without NiftyNet?

I didn't change test.py but the config file. At the line 272(The third line here): `print(all_vars)` `net1ax_vars = [x for x in all_vars if x.name[0:len(net_name1ax) + 1]==net_name1ax + '/']` `saver1ax...

训练过程中有一定概率因为label全为0而报错

**Describe the bug(问题描述)** 根据[这里](https://github.com/shenweichen/DeepCTR-Torch/blob/f6854257b1fc29caab655ce72c6b5528b7432574/deepctr_torch/models/basemodel.py#L264) 的代码，在verbose>0时，需要计算当前batch上的metric。而在特殊情况下，例如最后一个batch刚好只有几条label为0的数据时，会出现以下报错： > ValueError: y_true contains only one label (0.0). Please provide the true labels explicitly through the labels argument. 是否可以在trainer中添加让dataloader分布均匀的代码？