Liquid Visualize Evals Checkpoint

Visualize Evals Checkpoint

liquid-cookbooklanguage-modelslaptopedgeioscar-maker-identificationexamplesnotebooks.ipynb_checkpointslanguage-modelandroid

alph-notebooks/liquid-cookbook / visualize_evals-checkpoint.ipynb

Export

Run Notebooks

Contents

No cells yet

Add cells to see them here

Evaluation Results Visualization

This notebook loads and visualizes evaluation results from CSV files with embedded images.

[3]

The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

Print all evaluation samples

[16]

Number of classes: 36
Classes: ['Acura', 'Audi', 'BMW', 'Bentley', 'Bugatti', 'Buick', 'Cadillac', 'Chevrolet', 'Chrysler', 'Daewoo', 'Dodge', 'Ferrari', 'Fisker', 'Ford', 'GMC', 'HUMMER', 'Hyundai', 'Isuzu', 'Jeep', 'Lamborghini', 'Lincoln', 'MINI', 'Maybach', 'Mazda', 'Mercedes-Benz', 'Nissan', 'Plymouth', 'Porsche', 'Ram', 'Rolls-Royce', 'Spyker', 'Suzuki', 'Tesla', 'Toyota', 'Volkswagen', 'smart']
Total predictions: 100
Correct predictions: 63
Accuracy: 0.630

Calculate Accuracy

[13]

0.85

Print only misclassified samples

[14]

[ ]