fix: use exact 168-row lag for the seasonal-naive baseline
The baseline now reproduces mission/data.md exactly: MAE 5.40, WAPE 44.7%, R2 0.462. The model's lag_week remains the previous observed same weekday/hour, which handles sensor gaps and is documented in the model card.
This commit is contained in:
1 parent
939a8db482
commit
dc83825de0
9 files changed
+41
-26
No files matched your search
@@ -14,12 +14,12 @@
|
||||
"wape": 27.5,
|
||||
"r2": 0.777,
|
||||
"baseline": "seasonal_naive",
|
||||
"baseline_mae": 4.23,
|
||||
"baseline_wape": 35.0,
|
||||
"baseline_r2": 0.656,
|
||||
"baseline_mae": 5.4,
|
||||
"baseline_wape": 44.7,
|
||||
"baseline_r2": 0.462,
|
||||
"n_eval": 5350
|
||||
},
|
||||
"trained_at": "2026-11-14T04:00:08Z",
|
||||
"trained_at": "2026-11-14T04:03:37Z",
|
||||
"train_from": "2026-06-12T00:00:00Z",
|
||||
"train_to": "2026-11-07T23:00:00Z",
|
||||
"test_from": "2026-10-10T23:00:00Z",
|
||||
@@ -43,10 +43,10 @@
|
||||
"ablation": [
|
||||
{
|
||||
"name": "seasonal_naive",
|
||||
"description": "same local hour last week",
|
||||
"mae": 4.23,
|
||||
"wape": 35.0,
|
||||
"r2": 0.656,
|
||||
"description": "exact 168-row lag (same observed hour last week)",
|
||||
"mae": 5.4,
|
||||
"wape": 44.7,
|
||||
"r2": 0.462,
|
||||
"n": 5350
|
||||
},
|
||||
{
|
||||
|
||||
Reference in new issue
Block a user