=================================================================================

Some of the primary platforms for deep learning today are Summarized below:
         i) Theano (http://deeplearning.net/software/theano) is developed by the MILA lab at Université de Montréal,
         ii) TensorFlow (www.tensorflow.org) is developed by Google,
         iii) CNTK (https://github.com/Microsoft/CNTK) is developed by Microsoft.

Theano enables you to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. For a humongous volume of data, handcrafted C codes become slower. Theano enables swift implementations of code. Theano can recognise unstable expressions and yet compute them with stable algorithms which gives it an upper hand over NumPy. The closest Python package to Theano is Sympy.

The Keras handles the problem in a modular way as shown in Figure 4325a.

Figure 4325a. The deep-learning software and hardware stack in Keras process. [1]

============================================

[1] François Chollet, Deep Learning with Python, 2018.