titanic_workflow.bs

import brane_compute;
import brane_visualize;

// Function that mounts the train.csv and test.csv files in this brane package in the /data folder
print("### mounting the train.csv and test.csv file on /data ###");
print(mount());

// Function that creates /img folder for the visualizing graphs 
print("### creating /img folder for visualization graphs ###");
print(create_img());

// Function that returns the shape of the dataframe after reading the data from a file.
print("### Printing shape on train and test dataset ###");
print(shape("/data/train.csv"));
print(shape("/data/test.csv"));

// it first fetches the dataframe and calls the internal methods to plot graphs used for understanding the data (EDA) - plots histograms and bar plots 
print("### Running visualization function for EDA. 0 -> success, non-zero code -> error ###");
print(visualize_EDA("/data/train.csv"));

// Function that preprocesses the dataset (train, test) - 1) preprocesses the name feature to extract the title and create a different feature out of it. 2) imputs some logical value for the null/na value in a feature 3) changes categorical data to numerical data 4) handles the missing age for instances. 1 for training dataset and 0 for test dataset
print("### Preprocessing the dataset ###");
preprocess("/data/train.csv", 1);
preprocess("/data/test.csv", 0);


// Modelling and accuracy for random forest classifier
print("### Modelling and accuracy for  random forest classifier ###");
model("/data/prep_data1.csv", "/data/prep_data0.csv", "rfc");
print(accuracy("/data/prep_data1.csv", "rfc"));

// Modelling and accuracy for decision tree classifier
print("### Modelling and accuracy for  decision tree classifier ###");
model("/data/prep_data1.csv", "/data/prep_data0.csv", "dtc");
print(accuracy("/data/prep_data1.csv", "dtc"));

// Modelling and accuracy for Naive Bayes
print("### Modelling and accuracy for  Naive Bayes classifier ###");
model("/data/prep_data1.csv", "/data/prep_data0.csv", "bnb");
print(accuracy("/data/prep_data1.csv", "bnb"));

// Visualizing results for Random forest classifiers
print("### Running visualization function for results. 0 -> success, non-zero code -> error ###");
print(visualize_results("/data/prep_data1.csv", "rfc"));