Open serdarozsoy opened 7 years ago
I would like to see more deep diving in CNN:
Comparison of activation functions Creating right network architecture Techniques like Dropout, Batch-Normalization etc. Interactive applications like that: http://setosa.io/ev/image-kernels/
These concepts are explained in a great way:
5/20-Local Connectivity 6/20-Convolutional Layers 7/20-Strides and Padding 10/20-Pooling Layers 15/20-Image Augmentation 19/20- Transfer Learning
Keras practice parts could be more detailed with one by one explanation.
Distracting sounds almost never.