Can I install Tensorflow in Anaconda without using Keras? If I can what is the difference between using Keras with Tensorflow and only Tensorflow?
Can I install Tensorflow in Anaconda without using Keras?
If I can what is the difference between using Keras with Tensorflow and only Tensorflow?
While Tensorflow is a super powerful numerical computation and optimization library with lots of features for building neural networks, YET it is a bit tedious and nontrivial especially for beginners to use.
Here Keras comes very handy. In short, Keras is Tensorflow abstraction. It allows to quickly and easily define a neural net even complex ones with a few lines of codes. Also note that Keras not only runs on top of TensorFlow but also on top of Microsoft Cognitive Toolkit, Theano, or MXNet. Look at at this post where the blogger compares defining a typical neural network via directly in Tensorflow and Keras. You can see that with Tensorflow it is minimum 17 lines of code, whereas Keras reduce it to 10 lines of code.