d-li14 / face-attribute-prediction Goto Github PK

class fc_block(nn.Module):
def init(self, inplanes, planes, drop_rate=0.15):
super(fc_block, self).init()
self.fc = nn.Linear(inplanes, planes)
self.bn = nn.BatchNorm1d(planes)
if drop_rate > 0:
self.dropout = nn.Dropout(drop_rate)
self.relu = nn.ReLU(inplace=True)
self.drop_rate = drop_rate

def forward(self, x):
    x = self.fc(x)
    x = self.bn(x)
    if self.drop_rate > 0:
        x = self.dropout(x)
    x = self.relu(x)
    return x

self.bn = nn.BatchNorm1d(planes)
Expected more than 1 value per channel when training

E:\Users\Raytine\Anaconda3\python.exe F:/imageedit/face-attribute-prediction-master/main.py
=> creating model 'resnet18'
Traceback (most recent call last):
File "F:/imageedit/face-attribute-prediction-master/main.py", line 417, in
main()
File "F:/imageedit/face-attribute-prediction-master/main.py", line 135, in main
model = models.dictargs.arch
KeyError: 'resnet18'

could you provide the pre-trained model. Thanks a lot.

???? mails.tsinghua.edu.cn

There is no documentation on how to use it after training the model.

Thanks for your great implementation!

Can you please provide trained model weight?

Can you provide the pretrained model for test?

Hi, thanks for sharing your code! I also train this model. But the model seems like not converge fast as following.

I also use the img_align_celeba as training and val data and all training parameters used are default as your main.py. Could you know the reason why my model not converge?

It is a great work!

How do I need to crop and predict a custom image? ?

did you make it by yourself? what the format of it?

main.py imports "Bar", but there is no such definition.

tried many times, but still failed...

向以下收件人或组传递邮件失败:

[email protected] ([email protected])
你的邮件未送达。尽管反复尝试传递你的邮件，但是无法与远程服务器建立连接。

请通过其他方式(例如通过电话)与收件人联系，并让他们告诉电子邮件管理员，使其了解你的电子邮件系统貌似无法连接到电子邮件系统。向他们提供如下所示的错误详细信息。可能仅收件人的电子邮件管理员能够解决此问题。

有关详细信息和解决此问题的提示，请参阅此文章: https://go.microsoft.com/fwlink/?LinkId=389361。