THE RESULT
What the data shows
The largest computed metric is 0.3929 for Pastry -> Other_Faults; the smallest is 0 for Dirtiness -> Pastry. Metric: actual_class_share (Share of true-class test records).
Rates use the true-class denominator; the full CSV includes zero-count off-diagonal pairs.
| actual | predicted | count | actual class share | confusion |
|---|---|---|---|---|
| Bumps | Other_Faults | 33 | 0.3667 | Bumps -> Other_Faults |
| Other_Faults | Bumps | 28 | 0.2121 | Other_Faults -> Bumps |
| Pastry | Other_Faults | 11 | 0.3929 | Pastry -> Other_Faults |
| Other_Faults | Pastry | 6 | 0.04545 | Other_Faults -> Pastry |
| K_Scatch | Other_Faults | 5 | 0.05952 | K_Scatch -> Other_Faults |
| Z_Scratch | Other_Faults | 5 | 0.1471 | Z_Scratch -> Other_Faults |
| Dirtiness | Other_Faults | 3 | 0.3333 | Dirtiness -> Other_Faults |
| Bumps | Pastry | 3 | 0.03333 | Bumps -> Pastry |
| Other_Faults | Z_Scratch | 2 | 0.01515 | Other_Faults -> Z_Scratch |
| Bumps | Stains | 1 | 0.01111 | Bumps -> Stains |
| Other_Faults | Dirtiness | 1 | 0.007576 | Other_Faults -> Dirtiness |
| K_Scatch | Bumps | 1 | 0.0119 | K_Scatch -> Bumps |
THE METHOD
From source to answer
Row-normalized confusion table for the fixed forest on the grouped holdout.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
EVALUATION
How the result was checked
Fixed 80/20 split of unique input groups (seed 42); exact input duplicates stay together.
Fixed models; default decision threshold; no holdout tuning.
Target and input features
Target: fault
Inputs: pixels_areas, x_perimeter, y_perimeter, minimum_of_luminosity, maximum_of_luminosity, length_of_conveyer, typeofsteel_a300, typeofsteel_a400, steel_plate_thickness, edges_index, empty_index, square_index, outside_x_index, edges_x_index, edges_y_index, outside_global_index, logofareas, log_x_index, log_y_index, orientation_index, luminosity_index, sigmoidofareas
THE NEXT DECISION
What follows from the finding
Review recurring class confusions with inspectors before changing labels or workflow.
Where the conclusion stops
Every source row is a recorded fault. The data cannot estimate a production defect rate or distinguish healthy plates. Batch and machine IDs are absent, so grouped exact-input holdouts do not prove factory transfer.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.