Skip to content

End-to-End Notebooks

  • examples/notebooks/01_oil_discrimination_basic.ipynb
  • examples/notebooks/02_oil_vs_chips_matrix_effects.ipynb
  • examples/notebooks/03_hsi_surface_mapping.ipynb

Run all cells to see FoodSpec workflows end-to-end on synthetic data.

End-to-end notebooks

FoodSpec ships runnable notebooks under examples/notebooks/ to demonstrate full workflows with minimal setup.

Notebooks

  • 01_oil_discrimination_basic.ipynb
    Problem: oil authentication. Runs the oil discrimination protocol via Python/CLI, inspects figures/tables/narrative. Expected: confusion matrix, discriminative ratios, minimal panel.

  • 02_oil_vs_chips_matrix_effects.ipynb
    Problem: matrix effects. Uses oil + chips datasets to run the oil-vs-chips protocol; reviews divergence tables, effect sizes, and plots. Expected: matrix divergence tables/plots.

  • 03_hsi_surface_mapping.ipynb
    Problem: HSI segmentation/ROI analysis. Loads a synthetic HSI cube, runs segmentation → ROI → RQ, visualizes label maps and ROI spectra. Expected: label map, ROI spectra comparisons.

How to run

  1. Install (core is sufficient; HSI uses matplotlib/seaborn included): 
    pip install foodspec
  2. Launch Jupyter or VS Code: 
    python -m jupyter notebook
  3. Open the notebook and run all cells. Run folders are typically created under runs/; the notebook points to outputs.

What you should observe

  • Data loading (CSV/HDF5/HSI) and protocol selection via Python/CLI calls.
  • Bundle inspection (figures, tables, narrative).
  • Interpretation of key outputs: discriminative markers, matrix effects, heating trends, HSI segmentation.