facial-expression-classifier

DEMO: facial expression recognition with Keras model and pydantic settings

UV Setup

uv version
uv self update
uv python install 3.12
uv venv --python 3.12.6
uv python pin pypy@3.12
uv lock
uv sync --refresh
uv run pre-commit autoupdate
uv run pre-commit install --install-hooks

FER-2013 dataset

• consists of 48x48 pixel grayscale images of human faces
• automatically registered so each face is centered and occupies similar amount of space in each image
• contains 24400 images: with 22968 examples in training and 1432 examples in test
• one of seven categories (0=Angry, 1=Disgust, 2=Fear, 3=Happy, 4=Sad, 5=Surprise, 6=Neutral)

Train Model:

uv run python src/facial_expressions_classifier/run.py --epochs 10 --batch_size 30 --train_model

Plot Visualizations:

uv run ./src/facial_expressions_classifier/run.py --plot-history