How pluto13x/rock-paper-scissors-classification works
This is a small rock–paper–scissors image-classification experiment. Its workflow reads labeled images from the three category directories, resizes and binarizes them, extracts grid and line features, then splits the examples, fits a 3-neighbor classifier, and reports held-out accuracy. Feature calculations live in features.py; image binarization is implemented in funkcijenevezano.py. The README identifies a dataset and presentation, but gives no additional runtime wiring, so the graph reflects the sampled main.py workflow.
Main entry points: main.py Experiment Workflow · kamen Rock Images · papir Paper Images · makaze Scissors Images
Components 9
Labeled Images
- Rock Images
kamen - Paper Images
papir - Scissors Images
makaze
Image Processing
- Image Utilities
funkcijenevezano.py - Feature Extraction
features.py
Classification
- Experiment Workflow
main.py - Train Test Split
- KNN Classifier
- Accuracy Metric
Connections 8
- Experiment Workflow → Rock Images: reads images
main.py - Experiment Workflow → Paper Images: reads images
main.py - Experiment Workflow → Scissors Images: reads images
main.py - Experiment Workflow → Image Utilities: binarizes images
main.py - Experiment Workflow → Feature Extraction: extracts features
main.py - Experiment Workflow → Train Test Split: splits data
main.py - Experiment Workflow → KNN Classifier: fits and predicts
main.py - Experiment Workflow → Accuracy Metric: scores predictions
main.py