Build the first working regression loss workbench

Build only what the claim requires: both losses are computed from identical residuals and the outlier effect is visible. Prefer boring, deterministic code over frameworks you cannot yet explain. Run the path twice; identical output on this fixture is a feature, not a lack of creativity.

truth=[2,4,6,8]; pred=[2.5,3.5,6.5,12]
mse=sum((p-y)**2 for y,p in zip(truth,pred))/len(truth)
mae=sum(abs(p-y) for y,p in zip(truth,pred))/len(truth)
print(round(mse,3), round(mae,3))

Expected evidence: 2 for y,p in zip(truth,pred))/len(truth) mae=sum(abs(p-y) for y,p in zip(truth,pred))/len(truth) print(round(mse,3), .... Read each printed intermediate as part of the argument that the path works—not as decoration.

Narrate the journey from raw input to result for a single example from y_true/y_pred arrays including one large outlier. If you cannot name an intermediate, the implementation is still too opaque for this lab. Only after this path is solid should you generalize data sources or UI.

While implementing the regression loss workbench, print or log at least three intermediates that map to the claim (both losses are computed from identical residuals and the outlier effect is visible). Good intermediates are values a teammate could recompute with a calculator or diff. Bad intermediates are framework traces you cannot explain.

Re-run with y_true/y_pred arrays including one large outlier twice. If the second run differs, either the path is nondeterministic (document the seed) or you have hidden global state—both are lab bugs until named.

Stop adding features once the path supports compare MSE vs MAE on the same residual set to see which outlier hurts more. Extra UI, extra tools, or extra models belong in later chapters. The mastery bar for this page is simply: a deterministic end-to-end path with intermediates.

At the implementation stage for loss-functions, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about y_true/y_pred arrays including one large outlier that later pages inherit without redefining success. Keep that fixture small enough to inspect by hand, keep outputs copy-pasteable as text, and refuse to narrate this baseline as if it were a production SLA: mean prediction residual before any model.

For this page specifically, success looks like a deterministic path with printed intermediates while still centering the user decision to compare MSE vs MAE on the same residual set to see which outlier hurts more. If you cannot point to a file, command, or assertion that proves that for the regression loss workbench, stay on this page instead of advancing.

Glossary: loss function · Glossary: gradient descent

Previous · Next