lochan-eda / User Guide / AutomatedEDA

AutomatedEDA

High-level interface for the complete preprocessing workflow.

Class

from lochan_eda import AutomatedEDA

eda = AutomatedEDA()

Methods

prepare()

prepare( X: pd.DataFrame, target=None, exclude=None, split=True, test_size=0.2, random_state=42, stratify=None)

Provides a high-level interface for the common tabular preprocessing workflow.

Example

X_train, X_test, y_train, y_test = eda.prepare(
    df,
    target="target",
    exclude=["id", "name"],
    split=True,
    test_size=0.2,
    random_state=42,
    stratify=df["target"]
)

fit()

fit(X: pd.DataFrame, y: pd.Series=None)

Fits the preprocessing workflow using training data before it is reused for transformation.

Example

eda.fit(X_train, y_train)

transform()

transform(X: pd.DataFrame, y: pd.Series=None)

Applies the fitted preprocessing workflow to data.

Example

X_train, y_train = eda.transform(X_train, y_train)
X_test, y_test = eda.transform(X_test, y_test)

Train / test pattern

eda.fit(X_train)

X_train = eda.transform(X_train)
X_test = eda.transform(X_test)

← Profiler API Reference →