Frame the 3×3 plus-sign image classifier experiment

A lab needs a falsifiable claim before code. The claim here is that classify a tiny pixel grid as plus or other while keeping the pixels, score, threshold, and prediction visible. Record the tiny dataset, expected behavior, and one reason the result could be misleading. The first artifact is an experiment brief, not a model screenshot. It names the user, the decision the output supports, and the baseline you must beat. For this chapter, the baseline is deliberately transparent so later complexity has something honest to compare against.

The artifact's user-facing goal is specific: classify a tiny pixel grid as plus or other while keeping the pixels, score, threshold, and prediction visible. Its accepted input is exactly three rows of three binary brightness values. Those statements are intentionally narrower than “build an AI system.” Narrow scope lets us inspect every input and expected result, and it prevents a toy result from being presented as a production claim. Run the inventory below before implementing anything. Its output proves that the fixture is present and small enough to inspect by hand.

Save this as lesson.py and run python3 lesson.py. It uses only the language standard library, so the example is reproducible offline.

plus=[[0,1,0],[1,1,1],[0,1,0]]
print('pixels',sum(map(sum,plus)),'shape',len(plus),len(plus[0]))

Expected output: pixels 5 shape 3 3. Exact floating-point formatting may vary slightly, but the asserted behavior must not. Read the output as evidence about this stage, not merely proof that the interpreter started.

Print the 3×3 grid, flattened indices, five selected pixels, score, threshold, and label. If a rotated or shifted plus fails, that is an invariance limitation rather than a Python bug. Enumerate all nine single-pixel flips instead of trying one convenient corruption. Shape and binary-value errors should stop before scoring so malformed images cannot masquerade as confident predictions.

At the experiment brief stage, save the smallest failing fixture beside the expected result. Change one cause at a time and rerun the exact command printed above; that makes the repair reviewable and keeps this chapter's progressive artifact reproducible.

Use named positive and negative grids to compute confusion counts, then report every single-pixel-flip result. Accuracy alone hides whether the toy misses pluses or calls unrelated lines pluses, so include recall and false-positive count. Keep the threshold in the report and test a rotated plus separately. This evaluates a handcrafted detector, not a camera-ready vision model.

For this experiment brief page, preserve the fixture and result as evidence for the next page. Label observations separately from conclusions: a passing assertion establishes the behavior it names, while broader usefulness requires the chapter's full evaluation set and stated operating limits.