The main funcion is : Ml_Training.src.train.execute_all(model_repo, df_x, df_y, 'vine_model', savemodel=True) The function returns a dictionary with the model name, the best parameters and the f1 scores based on the
model_repo dictionary and the name of the saved model df_x: the dataframe without the class(column) we are to predict df_y: the part/class/column of the dataframe we predict filename: the filename we want to save the trained model savemodel:Bool - if True the function saves the model, if False it does not save
Feature of the MlTraining Package:
- Module for Data Preprocessing
- Module for Training, Testing
- Module for Data Validation
- Module for Serving the model