Notes on Python, machine learning and AI

Python520

The language itself: core features, data structures, OOP, testing, performance and the standard library.

Core language / Data structures and algorithms / Functional and itertools / OOP and design patterns / Performance and concurrency / Python for ML and AI / Scientific and math computing / Testing and best practices / Web APIs and automation

Machine learning501

Classical algorithms, model evaluation, feature work and the maths behind them.

Advanced ML topics / Classification and clustering / Dimensionality reduction and anomaly / Ensemble methods and boosting / Feature engineering and preprocessing / Fundamentals and overview / Model evaluation and validation / Regression / Regularization overfitting and optimization

Deep learning296

Neural networks from the ground up: layers, optimisers, CNNs, RNNs, GANs and training tricks.

Activation functions and loss / Autoencoders and generative / Backpropagation and optimization / CNN and computer vision / Fundamentals and architectures / KAN and frameworks / Neural network fundamentals / RNN LSTM and sequence / Regularization and training tricks

LLMs and generative AI166

Large language models, retrieval, agents, fine-tuning and prompting.

Agents and LangChain / Fine tuning and alignment / LLM applications and tools / LLM architecture and theory / RAG and retrieval / Vector databases and embeddings

NLP and transformers118

Text processing, embeddings, attention and transformer models.

BERT and language models / Fine tuning and adaptation / NLP core / Transformer architecture / Vision and multimodal

Statistics and maths138

Probability, statistics, linear algebra and calculus, with code.

Calculus and analysis / Linear algebra and matrices / Probability and distributions / Statistical inference / Time series and stochastic

Data science111

pandas, NumPy, exploratory analysis, cleaning and feature engineering.

Data visualisation69

Matplotlib, Seaborn and the craft of a readable chart.

Computer vision48

Images in, structure out: convolution, detection, segmentation, transfer learning.

Data engineering40

SQL, Spark, Kafka, pipelines and the plumbing under every model.

Web and APIs34

FastAPI, Flask, Django, REST and authentication.

MLOps and deployment17

Docker, CI/CD, serving models and tracking experiments.

Reinforcement learning9

Agents that learn by trying: Q-learning, policy gradients.

Miscellany215

Everything that did not fit elsewhere: tools, curiosities, side roads.

AI applications and systems / AI concepts and tools / Data engineering and tools / DataFrames and pandas / Embeddings and vectors / Software engineering / Statistics and analysis / Time series and sequences