5/2/2023 0 Comments Keras data augmentation![]() Note: in keras implementation convolution and dense layers have L2 kernel regularization, in pytorch implementation only the optimizer has L2. ![]() However, when I train this network on keras for 20 epochs, using the same data augmentation methods, I can reach over 70% validation accuracy. I know if the model’s capacity is low it is possible. I am training this network for 20 epochs, and I use the below data augmentation methods.Ĥ- random affine for horizontal and vertical translationĮach time I add a new data augmentation after normalization(4,5,6), my validation accuracy decreases from 60% to 50%. ![]() I have a cnn as below for cifar10: self.layer1 = nn.Sequential(
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |