DatAnalyzer.pipeline.regression module
- class DatAnalyzer.pipeline.regression.RegressionModel[source]
Bases:
LinearRegression,DecisionTreeClassifierThis class is used to create a regression model
- fit(X_train, y_train)[source]
Fit the data
- Parameters:
X_train (pandas.DataFrame) – Train data
y_train (pandas.DataFrame) – Target
- Returns:
self
- Return type:
- predict(X_test)[source]
Predict the data
- Parameters:
X_test (pandas.DataFrame) – Test data
- Returns:
Predicted data
- Return type:
pandas.DataFrame
- regression(model)[source]
Return the regression model
- Parameters:
model (str) – Type of regression
- Returns:
Regression model
- Return type:
sklearn.linear_model.LinearRegression
- Raises:
ValueError – If the model is not supported
- regression_results(y_true, y_pred)[source]
Return the regression metrics
- Parameters:
y_true (pandas.DataFrame) – True values
y_pred (pandas.DataFrame) – Predicted values
- Returns:
Regression metrics
- Return type:
dict
- split_data(data, dependant_variable_column, random_state, test_size)[source]
Split the data into train and test data
- Parameters:
data (pandas.DataFrame) – Data to split
dependant_variable_column (str) – Name of the dependant variable column
random_state (int) – Random state
test_size (float) – Size of the test data
- Returns:
Train and test data
- Return type:
tuple