RegressorChain

A multi-output model that arranges regressors into a chain.

This will create one model per output. The prediction of the first output will be used as a feature in the second output. The prediction for the second output will be used as a feature for the third, etc. This "chain model" is therefore capable of capturing dependencies between outputs.

Parameters

• model

  Type → base.Regressor

  The regression model used to make predictions for each target.

• order

  Type → list | None

  Default → None

  A list with the targets order in which to construct the chain. If None then the order will be inferred from the order of the keys in the target.

Examples

from river import evaluate
from river import linear_model
from river import metrics
from river import multioutput
from river import preprocessing
from river import stream

from sklearn import datasets

dataset = stream.iter_sklearn_dataset(
    dataset=datasets.load_linnerud(),
    shuffle=True,
    seed=42
)

model = multioutput.RegressorChain(
    model=(
        preprocessing.StandardScaler() |
        linear_model.LinearRegression(intercept_lr=0.3)
    ),
    order=[0, 1, 2]
)

metric = metrics.multioutput.MicroAverage(metrics.MAE())

evaluate.progressive_val_score(dataset, model, metric)

MicroAverage(MAE): 12.649592

Methods

learn_one

Fits to a set of features x and a real-valued target y.

Parameters

• x
• y
• kwargs

Returns

self

predict_one

Predict the outputs of features x.

Parameters

• x
• kwargs

Returns

The predictions.