smiler
smiler
> In keras_transfer_cifar10.ipynb > When you train a shallow net > > ``` > model.add(Conv2D(filters=100, kernel_size=2, input_shape=features.shape[1:])) > ``` > features since they come from bottle neck, they are of...
thanks for your good guys!
i run this in win10
> Hi, @smiler96 > I use the command provided by author(using pre-trained RCAN model) and meet the same problem. > So, you use pre-trained RCAN.pt? It seems that train the...
> thanks @smiler96 , but have you trained the HAN? how about its final result on benchmark? I merge HAN into EDSR-pytorch repo(because my GPU can't support cuda8) and the...
> > > Hi, @smiler96 > > > I use the command provided by author(using pre-trained RCAN model) and meet the same problem. > > > So, you use pre-trained...
> > Hi, @smiler96 I use the command provided by author(using pre-trained RCAN model) and meet the same problem. So, you use pre-trained RCAN.pt? It seems that train the whole...
> Very nice job!!! I tried to use resnet (10, 12, 18) as a backbone of my model, but it didn't improve performance as we expected. In contrast, In our...
> Pretraining the backbone on the train dataset can effectively avoid overfitting. pretraining is everything!
> Hello,I haven't changed the parameters. The network precision is 76,releationnet+resnet18 hello, i train renet10+protnet while it overfits after 10k episodes, have you met this? no data augmentation in training...