Theano

What do you like best?

The level of control to write algorithms/networks from scratch is relatively high. It also is faster in executing the networks and comes with an optional keras frontend to make it easier for developer to develop networks.

What do you dislike?

Without keras as frontend, using raw theano has a high learning curve.

Recommendations to others considering the product:

Initially, It might be hard to figure out how things work with theano but once picked up, you will find it more stable and optimized. So its worth the initial effort

What problems are you solving with the product? What benefits have you realized?

It was harder for me to develop networks from scratch because it has a steeper learning curve. When I figured using it eventually, then I found it more stable and fast compared to other deep learning frame works.