User defined model

As it’s described in the definition of functions, the user-defined models can be accepted as an input to be trained and evaluated in a package functions. The format of these functions must be in accordance with the following description:

Parameters

#	Input Name	Input Description
1	X_training	type: dataframe details: the dataframe of features in the training set which is used to train the models.
2	X_validation	type: dataframe details: the dataframe of features in the validation set which is used to evaluate models’ performance.
3	Y_training	type: array details: The values of the target variable in the training set.

Returns

#	Output Name	Output Description
1	train_predictions	type: list details: The predicted values of target variable in the training part
2	validation_predictions	type: list details: The predicted values of target variable in the validation part
3	trained_model	type: object details: The trained model on the the training part

The custom model function must be defined in this format:

def custom_model_name(X_training, X_validation, Y_training):
     # import required packages
     # define your model
     # train the model on X_training and Y_training
     # predict values on X_training
     # predict values on X_validation
     return(train_predictions, validation_predictions, trained_model)