Table of Contents

• Getting started

  • Overview: the machine learning process

  • Pytorch & other tools, installation

  • Project: Fashion MNIST

• Tensors

  • What's a tensor?

  • Operations on tensors

  • Tensor calculus

  • Project: Images

• Differentiation

  • Differential calculus

    • Derivative, partial derivative

    • Gradient, Jacobian Matrix,

    • Differential, Chain rule

    • higher order differential calculus

  • Differentiable programming:

    • Automatic differentiation

    • Computation graph

    • Forward/backward differentiation

    • In Pytorch

    • Tensor, Variable, Function

  • Project: Hamiltonian mechanics (?)

• Optimization

  • Concepts

  • Torch optimizers

  • Project: Quadratic function / Rosenbrock function / Map (?)

• Datasets

  • Tables & pandas

  • ... (structure & formats & tools "in the wild"?)

• Machine learning

  • Linear models

  • Logistic regression

  • Neural networks

References

• 📖 🇫🇷 Calcul Différentiel, Intégrale et Stochastique by Sébastien Boisgérault, Thomas Romary, Emilie Chautru et Pauline Bernard.

• 📖 🇺🇸 Elements of Differentiable Programming by Mathieu Blondel and Vincent Roulet.

• 📖 🇺🇸 The Little Book of Deep Learning by François Fleuret.

• 📖 🇺🇸 Learning Theory from First Principles by Francis Bach.

• 📖 🇺🇸 Scientific Visualization: Python + Matplotlib by Nicolas Rougier.