I also recommend reading @fchollet's guide on integrating Keras and TensorFlow, especially for those wanting to implement novel components at a lower level :) http://blog.keras.io/keras-as-a-simplified-interface-to-tens...

I second that. Keras was the first really intuitive interface to neural networks that I stumbled upon. It was also convenient that it was built in Python and played well with the numpy/scipy ecosystem. Thank you fchollet!

Thanks for posting that - keras looks a lot easier to use for mere amateurs like myself.

Keras is the best framework out there.

A simple python layer is exactly what I was looking for. Thanks!