AI
R
Python
Julia
Git
Bash/Zsh
Emacs
Tools
Deep learning experiments tracking with MLflow
AI
Intro DL with PyTorch
Introduction
Which framework to choose?
High-level frameworks
Introduction to NN
The PyTorch API
PyTorch tensors
Automatic differentiation
Workflow
Creating checkpoints
Example: the MNIST
DL on production clusters
Resources
The JAX library
Why JAX?
How does it work?
Installation
Running jobs
Relation to NumPy
JIT compilation
Benchmarking JAX code
Accelerators
Pytrees
Automatic differentiation
Parallel computing
Pushing optimizations further
Libraries built on JAX
Resources
Intro DL with JAX
Introduction: JAX
Running JAX
Installing packages
Relation to NumPy
Loading data
Preprocessing data
Defining a model architecture
Loading pre-trained weights
Hyperparameters
Fine-tuning the model
Training at scale
Resources
ML with Scikit-learn
What is scikit-learn?
Sklearn workflow
Workshops
Audio DataLoader with PyTorch
Finding pre-trained models
Intro ML for the humanities
Quick intro to DL, NLP, and LLMs
Webinars
Map of current ML frameworks
Data & model version control
Accelerated array & AD with JAX
Image upscaling
AI-powered coding with Copilot
Easier PyTorch with fastai
DL in Julia with Flux
PyTorch tensors in depth
Bayesian inference in JAX
Experiments tracking
Deep learning experiments tracking with MLflow
Author
Marie-Hélène Burle
Coming up on November 25.
Bayesian inference in JAX