TianshengSun

Hi, i have noticed that there is a ```python rgb_fc_out = tf.layers.dense(rgb_logits_dropout, _CLASS_NUM[dataset], use_bias=True) ``` in your test.py. I don't understand about this. This layer looks like have no pretrained...

The link of pretrained model is not availbale？？？

2

It seems like the link of pretrained model is not availbale. So , where can i find the pretrained model???

Hi, Thanks for this implementation. But I still have some problems. The parameters in model are initialize as follow： ```python def _init_weights(self, m): if isinstance(m, nn.Linear): with torch.no_grad(): trunc_normal_(m.weight, std=.02)...

Hi, Thanks for this implementation. I saw the parameters of nn.Linear() are set to no_gard() in models.py Line:139. ``` python @torch.no_grad() def init_weights(self): def _init(m): if isinstance(m, nn.Linear): nn.init.xavier_uniform_(m.weight) #...

机器配置问题

7

请问我可以用GTX1650 4G 8G运行内存的笔记本运行该程序吗，为什么我设置成GPU运行在训练的刚开始的时候就会导致运行内存很快达到8G然后进程被自动结束？

Hi, I want use Inceptionv3 in my network, but i only want use some layers in Inceptionv3. I was trying to use tensorflow_hub to load model , but it can...