在我们的示例中,sktime 的方法 make_reduction() 使用 scikit-learn 的模型基于约简创建预测器。它接受一个回归量,即预测策略的名称和窗口长度。它输出一个可以像任何其他预测器一样安装的预测器。您也可以使用 DirectTabularRegressionForecaster 对象将预测问题简化为表格回归任务。但是,该预测器使用直接减少策略。值...
When using an sklearn model, that does not support NaN values withmake_reductionand ayseries with NaN values, the resulting exception is misleading: ValueError: Input X contains NaN. To Reproduce importnumpyasnpfromsklearn.linear_modelimportLinearRegressionfromsktime.datasetsimportload_longleyfromsktime...
sktime is composable and compatible with sklearn – and any sklearn compatible estimator can be used, e.g. in reduction compositors: import xgboost as xgb xgbregressor = xgb.XGBRegressor() forecaster = make_reduction( xgbregressor, window_length=15, strategy="recursive") ...
Makefile Rearchitecting skpro on skbase - part 1: CI-CD, base classes, pac… May 12, 2023 README.md Release 2.4.1 (#411) Jun 27, 2024 conftest.py [MNT] Differential testing of estimators (#96) Sep 13, 2023 pyproject.toml [MNT] [Dependabot](deps): Update polars requirement from...
When using an sklearn model, that does not support NaN values with make_reduction and a y series with NaN values, the resulting exception is misleading: ValueError: Input X contains NaN. To Reproduce import numpy as np from sklearn.linear_model import LinearRegression from sktime.datasets impor...