CNN Intuition and Transfer Learning
By the end of this module you can explain what a convolutional network learns layer by layer, and use transfer learning to get a working classifier from a small dataset without pretending you trained it from scratch.
Units
- Unit 06.00: What a convolution actually does to a patch
- Unit 06.01: Why transfer learning works, and when it does not
- Unit 06.02: Freezing, unfreezing and the learning rate
- Unit 06.03: Preprocessing that must match the pretrained model
- Unit 06.04: A baseline you must beat before any of this
