index.html

<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Tiny Cats Model — Classifier Demo</title>
<style>
  body { font-family: system-ui, sans-serif; max-width: 760px; margin: 3rem auto;
        padding: 0 1rem; color: #1f2937; line-height: 1.55; }
  h1 { font-size: 1.6rem; }
  table { border-collapse: collapse; margin: 1rem 0; width: 100%; }
  th, td { border: 1px solid #d1d5db; padding: .5rem .75rem; text-align: left; }
  th { background: #eff6ff; }
  a { color: #1d4ed8; }
  .muted { color: #6b7280; font-size: .85rem; }
</style>
</head>
<body>
<h1>Tiny Cats Model — Classifier Demo</h1>
<p>Binary <strong>cat vs dog</strong> classifier (ResNet-18, 1478 images per
split eval, Oxford IIIT Pet), trained with class-balanced loss on the full
dataset. Evaluation report in <code>evaluation_report.json</code> at the
<a href="https://github.com/d-oit/tiny-cats-model">repo</a> or the
<a href="https://huggingface.co/d-oit/tiny-cats-model">model hub repo</a>.</p>

<h2>Headline metrics</h2>
<table>
<tr><th>Metric</th><th>Value</th></tr>
<tr><td>Accuracy</td><td>99.66% (1473/1478)</td></tr>
<tr><td>Macro F1</td><td>99.61%</td></tr>
<tr><td>Weighted F1</td><td>99.66%</td></tr>
<tr><td>Misclassifications</td><td>5</td></tr>
</table>

<h2>Per-class precision / recall / F1</h2>
<table>
<tr><th>Class</th><th>Precision</th><th>Recall</th><th>F1</th></tr>
<tr><td>cat</td><td>99.58%</td><td>99.37%</td><td>99.48%</td></tr>
<tr><td>other</td><td>99.70%</td><td>99.80%</td><td>99.75%</td></tr>
</table>

<h2>Confusion matrix (rows=true, cols=predicted)</h2>
<table>
<tr><th></th><th>cat</th><th>other</th></tr><tr><th>cat</th><td>476</td><td>3</td></tr><tr><th>other</th><td>2</td><td>997</td></tr>
</table>

<p class="muted">Report timestamp: 2026-08-12T10:51:31.480493+00:00</p>
</body>
</html>