Overview of traditional ML algorithms

https://ataiva.com/a-deep-dive-into-machine-learning-algorithms/

With scikit

  • Linear regression
  • Logistic regression
  • Naive Bayes

Select best ML classifier

github.com/shankarpandala/lazypredict

ML and AI primers

http://primers.aman.ai

Deep Learning: A crash course

https://www.youtube.com/watch?v=r0Ogt-q956I

Great broad intro

  • 3 classes : supervised, unsupervised,
  • Perceptron, model of a neuron
  • Activation functions
  • Training and testing set
  • Validation with small sets (images of Pluto): cross-correlation validation
  • Convolution

Before you start

https://eugeneyan.com/writing/first-rule-of-ml/

  • before applying ML to a problem, start with simpler heuristics
  • get to know the data, visualize relations

Understanding auto differentiation

https://vmartin.fr/understanding-automatic-differentiation-in-30-lines-of-python.html

Models, methods

https://paperswithcode.com/methods

Datasets

https://paperswithcode.com/datasets

#PyTorch or TensorFlow ?

https://www.assemblyai.com/blog/pytorch-vs-tensorflow-in-2022/

PyTorch Lightning

https://www.assemblyai.com/blog/pytorch-lightning-for-dummies/

Forecasting time series

https://www.bounteous.com/insights/2020/09/15/forecasting-time-series-model-using-python-part-one/

Tools / environments

Jupyter Notebooks

RISE, tool to transform Jupyter notebooks into a slide show

Jupyter notebook extensions

Useful ones:

  • Table of Contents

Conda