A collection of tutorials, in the form of Jupyter notebooks, to help people getting started with Machine Learning in Python. Some emphasis is put on the theoretical aspect behind what the code does and sometimes tasks are performed first using low-level tools (e.g. ScyPy) and then with more high-level ones (e.g. SKLearn).
- Linear regression and binary classification.
- Neural networks: using Keras layers with TensorFlow (code taken from https://blog.keras.io/keras-as-a-simplified-interface-to-tensorflow-tutorial.html, with some explanation around it).
- Neural networks: (nonlinear) regression with Keras.
Keras is now used as part of Tensorflow, not as a separate library. We only use a deprecated way of loading the MNIST dataset, as in the new one the next_batch function is not implemented anymore.