tier 3 · fault_location · line view · regression · all cells measured
Headline metric mae (↓ lower is better); full metric set: mae · rmse · r2 · median_abs_error. Metrics are a frozen contract (src/evemtbench/evaluation/metrics.py); label derivation: src/evemtbench/tasks/labels.
event flt_1phg_shc · grid double_line (adapt_grid/test split) · sample 23, window #529, cubicle 3 (max-RMS) · 50 ms / 480 samples @ 9.6 kHz · top: 3-phase current, bottom: 3-phase voltage · figure provenance in assets/manifest.json
| task | grid | baseline | headline | test | benchmark | params | fit/seed | trace |
|---|---|---|---|---|---|---|---|---|
| fault_location_line | cigre_mv | majority | mae ↓ | 23.915 ± 0.000 (n=5) | 30.351 ± 0.000 (n=5) | — | 0s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | cigre_mv | random_forest | mae ↓ | 24.443 ± 0.038 (n=5) | 31.076 ± 0.051 (n=5) | — | 1.2m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | cigre_mv | mlp | mae ↓ | 23.915 ± 0.057 (n=5) | 30.470 ± 0.095 (n=5) | 3.1M | 21s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | cigre_mv | gru | mae ↓ | 23.887 ± 0.041 (n=5) | 30.485 ± 0.062 (n=5) | 55k | 39s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | cigre_mv | cnn | mae ↓ | 23.005 ± 0.552 (n=5) | 24.522 ± 4.373 (n=5) | 51k | 1.1m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | cigre_mv | resnet | mae ↓ | 23.890 ± 0.162 (n=5) | 29.672 ± 2.190 (n=5) | 509k | 1.9m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | majority | mae ↓ | 24.416 ± 0.000 (n=5) | 31.094 ± 0.000 (n=5) | — | 0s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | random_forest | mae ↓ | 19.120 ± 0.036 (n=5) | 14.992 ± 0.059 (n=5) | — | 1.2m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | mlp | mae ↓ | 16.095 ± 0.030 (n=5) | 5.179 ± 0.243 (n=5) | 3.1M | 1.1m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | gru | mae ↓ | 16.546 ± 0.127 (n=5) | 6.248 ± 0.522 (n=5) | 55k | 46s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | cnn | mae ↓ | 17.841 ± 0.524 (n=5) | 10.421 ± 1.154 (n=5) | 51k | 1.1m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | double_line | resnet | mae ↓ | 17.199 ± 0.263 (n=5) | 8.809 ± 1.116 (n=5) | 509k | 3.7m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | majority | mae ↓ | 24.378 ± 0.000 (n=5) | 31.168 ± 0.000 (n=5) | — | 0s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | random_forest | mae ↓ | 21.576 ± 0.051 (n=5) | 19.552 ± 0.089 (n=5) | — | 59s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | mlp | mae ↓ | 17.141 ± 0.065 (n=5) | 6.545 ± 0.540 (n=5) | 3.1M | 59s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | gru | mae ↓ | 17.256 ± 0.294 (n=5) | 6.415 ± 0.903 (n=5) | 55k | 1.0m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | cnn | mae ↓ | 17.693 ± 0.366 (n=5) | 8.842 ± 1.200 (n=5) | 51k | 1.6m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | testgrid_110kv | resnet | mae ↓ | 17.864 ± 0.278 (n=5) | 8.733 ± 0.828 (n=5) | 509k | 3.1m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | majority | mae ↓ | 25.224 ± 0.000 (n=5) | 31.129 ± 0.000 (n=5) | — | 0s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | random_forest | mae ↓ | 23.590 ± 0.020 (n=5) | 23.631 ± 0.039 (n=5) | — | 1.2m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | mlp | mae ↓ | 18.681 ± 0.137 (n=5) | 9.396 ± 0.257 (n=5) | 3.1M | 1.0m | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | gru | mae ↓ | 18.887 ± 0.218 (n=5) | 10.259 ± 0.662 (n=5) | 55k | 53s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | cnn | mae ↓ | 20.081 ± 0.231 (n=5) | 13.893 ± 0.551 (n=5) | 51k | 40s | trace: output · eval-code · log · train-code · config · W&B |
| fault_location_line | ieee39 | resnet | mae ↓ | 20.500 ± 0.473 (n=5) | 15.229 ± 0.959 (n=5) | 509k | 1.8m | trace: output · eval-code · log · train-code · config · W&B |
headline & reported metrics for this view — the metric set is a frozen contract; each links to the exact function that computes it
| metric | definition | entry |
|---|---|---|
mae | mean absolute error, in the grid's own distance units | evaluate() |
rmse | root mean squared error | evaluate() |
r2 | coefficient of determination 1 − SS_res/SS_tot | evaluate() |
median_abs_error | median of absolute errors | evaluate() |
PYTHONPATH=src python -m evemtbench.baselines.runner --task fault_location_line --protocol held_out \
--grid <grid> --baseline <baseline> --seeds 0 1 2 3 4