JiaqingFu

I run the train_speech_embedder.py with default config, but I get the result : EER 0.0723 with final checkpoint ,get the result :EER 0.0578 with intermediate checkpoint @HarryVolek @seandickert @mazzzystar @pranshurastogi29

my code ``` class batch_norm_2d(nn.Module): def __init__(self,num_features,eps=1e-5,momentum=0.1): super(batch_norm_2d,self).__init__() self.bn = nn.BatchNorm3d(num_features) def forward(self,input): y = input.transpose(0,2).contiguous().transpose(0,1).contiguous() y = self.bn(y,4) return y.contiguous().transpose(0,1).contiguous().transpose(0,2) ``` the error information ``` Traceback (most recent call...

there is a wrong with parameter alpha of leakyrelu @traveller59 @FindDefinition @riyadshairi979 @Benzlxs @xmyqsh

未来需要合并的算法

2

1.ghostnet，参考链接：https://pytorch.org/hub/pytorch_vision_ghostnet/ 2.行为识别（TSN），参考链接：https://github.com/open-mmlab/mmaction/blob/master/mmaction/models/recognizers/TSN2D.py 3.人脸分割，参考链接：https://github.com/nasir6/face-segmentation 4.超分辨率 （SRGAN），参考链接：https://arxiv.org/pdf/1609.04802v1.pdf 5.REID（还在开发中）

Is the train data available

2

I want to obtain the data to train the model on [oneflow](https://github.com/Oneflow-Inc/oneflow)（a new open-source deep learning framework），3Q！

When trying to use model.save_weights,get the h5py name already exist error

9

The model I used is a simple seq2seq attention model, When I try to save the model weight after 5 epoch, I get the following error, It is likely the...