Machine Learning

Beginner ML projects for learning by running, inspecting, modifying, and rebuilding.

Setup

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Projects

Run the Titanic survival classifier:

python 01_titanic_survival/train.py

Run the house price regression model:

python 02_house_price_regression/train.py

Run the spam detector:

python 03_spam_detector/train.py

Run the tiny neural network from scratch:

python 04_tiny_neural_network/train.py

Run the PyTorch neural network:

python 05_pytorch_neural_network/train.py

Run NumPy linear regression from scratch:

python 06_linear_regression_from_scratch/train.py

Run NumPy logistic regression from scratch:

python 07_logistic_regression_from_scratch/train.py

Run the MNIST digit classifier:

python 08_mnist_digit_classification/train.py

Open the EDA notebook:

09_data_processing_and_eda/eda_titanic.ipynb

Run the CIFAR-10 CNN:

python 10_cifar10_cnn/train.py --fake-data --epochs 1 --max-train-batches 5 --max-test-batches 2

Run the embeddings text classifier:

python 11_embeddings_text_classifier/train.py

Run the tiny transformer language model:

python 12_tiny_transformer_language_model/train.py --max-steps 20 --eval-interval 10

Study Loop

1. Run one project.
2. Ask AI to explain the code section by section.
3. Add print statements to inspect the data.
4. Change one thing and rerun it.
5. Rewrite the skeleton in a new file from memory.