Given that Keras is a high level API and can sit on top of machine learning libraries (such as Tensorflow) - then it seems like a logical choice to adopt it.
I referenced TFLearn a while ago - that’s an abstraction of Tensorflow - primarily aimed at helping you learn the ropes - don’t think it was intended for anything further that.
So now we’re getting a bit more serious - it’s time to invest some time learning how to use Keras.
There’s a ton of material to get you started with Keras: Machine Learning Mastery is a great site. Jason Brownlee has selflessly developed the site and it’s got some awesome examples to get you started.
The Keras website itself has a load of examples in the documentation section too - I just found Jason’s site a little easier to digest.
OK here goes …