preprocessing.exog_providers.EntsoeForecastLoadProvider

preprocessing.exog_providers.EntsoeForecastLoadProvider(
    data_home=None,
    max_gap=0,
    max_tail_gap=0,
    provider_window=None,
)

ENTSO-E day-ahead Forecasted Load as an exogenous near-oracle prior.

Wraps spotforecast2_safe.data.fetch_data.load_timeseries_forecast, which reads the Forecasted Load column already merged into interim/energy_load.csv. The day-ahead forecast is published on D-1 and is therefore genuinely available at forecast time (leakage-clean, CR-3).

Parameters

Name	Type	Description	Default
data_home	`DataHome`	Root data directory forwarded to the loader. `None` resolves via `get_data_home()`.	`None`
max_gap	int	Maximum contiguous missing-value run healed by `_align_to_index`. See :func:`_align_to_index` for full semantics. Defaults to `0`.	`0`
max_tail_gap	int	Extended healing budget for the trailing-edge NaN run. See :func:`_align_to_index`. Defaults to `0`.	`0`
provider_window	Optional[pd.DatetimeIndex]	Validation index passed to `_align_to_index` as validate_index. See :func:`_align_to_index`. Defaults to `None`.	`None`

Examples

import os
import shutil
import tempfile

import pandas as pd

from spotforecast2_safe.preprocessing.exog_providers import (
    EntsoeForecastLoadProvider,
)

tmp = tempfile.mkdtemp()
os.environ["SPOTFORECAST2_DATA"] = tmp
interim = os.path.join(tmp, "interim")
os.makedirs(interim, exist_ok=True)
idx = pd.date_range("2023-01-01", periods=48, freq="h", tz="UTC")
pd.DataFrame(
    {"Actual Load": 100.0, "Forecasted Load": 99.0}, index=idx
).rename_axis("Time (UTC)").to_csv(os.path.join(interim, "energy_load.csv"))

out = EntsoeForecastLoadProvider().build(idx)
print(out.columns.tolist(), out.shape)

shutil.rmtree(tmp)
del os.environ["SPOTFORECAST2_DATA"]

['entsoe_forecasted_load'] (48, 1)

Methods

Name	Description
build	Return the day-ahead Forecasted Load aligned to index.

build

preprocessing.exog_providers.EntsoeForecastLoadProvider.build(index)

Return the day-ahead Forecasted Load aligned to index.

Parameters

Name	Type	Description	Default
index	pd.DatetimeIndex	Hourly `DatetimeIndex` (tz-aware UTC) for the forecast window.	required

Returns

Name	Type	Description
	pd.DataFrame	pd.DataFrame: Single column `entsoe_forecasted_load`, `float32`.

Raises

Name	Type	Description
	`ExogProviderError`	If the interim CSV is missing or the `Forecasted Load` column is absent.

Examples

import os
import shutil
import tempfile

import pandas as pd

from spotforecast2_safe.preprocessing.exog_providers import (
    EntsoeForecastLoadProvider,
)

tmp = tempfile.mkdtemp()
os.environ["SPOTFORECAST2_DATA"] = tmp
os.makedirs(os.path.join(tmp, "interim"), exist_ok=True)
idx = pd.date_range("2023-06-01", periods=24, freq="h", tz="UTC")
pd.DataFrame(
    {"Actual Load": 100.0, "Forecasted Load": 98.0}, index=idx
).rename_axis("Time (UTC)").to_csv(
    os.path.join(tmp, "interim", "energy_load.csv")
)

out = EntsoeForecastLoadProvider().build(idx)
print(out.columns.tolist(), out.shape, out.dtypes.iloc[0].name)
assert out.shape == (24, 1)

shutil.rmtree(tmp)
del os.environ["SPOTFORECAST2_DATA"]

['entsoe_forecasted_load'] (24, 1) float32