Deep learning using PyTorch - Transfer Learning

There are 2 common myths about deep learning for Computer Vision tasks.

1. You need so much data.
2. Its too expensive , it needs a lot of computation power.

It turns out that both of them are just myths when handled in the right way. Using a concept called transfer learning , you can build state of art classifiers in less than a few dollars and less than an hour. Sounds interesting check out my blogs about transfer learning for tricks that are used across organisations like Facebook .