Python libraries for ML
Although you can implement all ML algorithms using almost any programming language, there are many Python 3 modules that can help you work with ML without having to run everything from scratch.
The scikit-learn module is built on top of NumPy, SciPy and matplotlib, and offers algorithms for Classification, Regression, Clustering, Model selection, Dimensionality reduction and data Preprocessing. Knowing more about Numpy, matplotlib or Scipy modules with help you, especially if you want to plot any data. The TensorFlow library is also ideal for maths operations such as performing operations with multidimensional arrays as well as vectors and matrices. There’s also a lighter version of TensorFlow called TensorFlow Lite for mobile and embedded devices. You can find more information about TensorFlow Lite at https://github.com/ tensorflow/tensorflow/tree/master/tensorflow/contrib/lite.
Another popular library is the Keras Python Deep Learning library, which runs on top of TensorFlow, CNTK or Theano. You can learn more about Keras at https://keras.io. CNTK is an open source, deeplearning toolkit developed by Microsoft – find out more about CNTK at https://github.com/Microsoft/cntk.
Finally, Theano is a Python library that can help you work with mathematical expressions that contain multi-dimensional arrays ( numpy.ndarray). Additionally, Theano can be used for Deep Learning in Python because it enables you to create Deep Learning models. See more at https://github.com/Theano/Theano.