File size: 2,657 Bytes
47d42b1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 | """Artifact saving utilities for train_xgboost."""
from __future__ import annotations
import json
import logging
from pathlib import Path
import joblib
import pandas as pd
from xgboost import XGBRegressor
LOGGER = logging.getLogger("static_xgboost")
def save_artifacts(
model: XGBRegressor,
metrics: dict[str, float],
pred_df: pd.DataFrame,
x_columns: list[str],
artifacts_dir: Path,
target_mode: str,
) -> None:
"""Save model, metrics, feature importance, and holdout predictions."""
artifacts_dir.mkdir(parents=True, exist_ok=True)
suffix = "ghi" if target_mode == "ghi" else "temp"
model_path = artifacts_dir / f"xgboost_{suffix}_model.joblib"
metrics_path = artifacts_dir / f"xgboost_{suffix}_metrics.json"
predictions_path = artifacts_dir / f"xgboost_{suffix}_test_predictions.csv"
feature_importance_path = artifacts_dir / f"xgboost_{suffix}_feature_importance.csv"
joblib.dump(model, model_path)
payload = {
"model_name": f"xgboost_{target_mode}_next_1h",
"target_mode": target_mode,
"feature_count": len(x_columns),
"features": x_columns,
**metrics,
}
metrics_path.write_text(json.dumps(payload, indent=2), encoding="utf-8")
pred_df.to_csv(predictions_path, index=False)
importance = pd.DataFrame(
{
"feature": x_columns,
"importance": model.feature_importances_,
}
).sort_values("importance", ascending=False)
importance.to_csv(feature_importance_path, index=False)
LOGGER.info("Saved model: %s", model_path)
LOGGER.info("Saved metrics: %s", metrics_path)
LOGGER.info("Saved predictions: %s", predictions_path)
LOGGER.info("Saved feature importance: %s", feature_importance_path)
def save_rolling_backtest_artifacts(
folds_df: pd.DataFrame,
summary: dict[str, float],
artifacts_dir: Path,
target_mode: str,
) -> None:
"""Save rolling backtest fold metrics and summary."""
artifacts_dir.mkdir(parents=True, exist_ok=True)
suffix = "ghi" if target_mode == "ghi" else "temp"
folds_path = artifacts_dir / f"xgboost_{suffix}_rolling_backtest_folds.csv"
summary_path = artifacts_dir / f"xgboost_{suffix}_rolling_backtest_summary.json"
folds_df.to_csv(folds_path, index=False)
summary_payload = {
"model_name": f"xgboost_{target_mode}_next_1h",
"target_mode": target_mode,
**summary,
}
summary_path.write_text(json.dumps(summary_payload, indent=2), encoding="utf-8")
LOGGER.info("Saved rolling folds: %s", folds_path)
LOGGER.info("Saved rolling summary: %s", summary_path)
|