/Titanic-Machine-Learning

Intro EDA and machine learning decision tree model to predict surviving the Titanic disaster.

Primary LanguageJupyter Notebook

Titanic Machine Learning Introduction

The following notebook walks through basic EDA and then deploys a machine learning model to correctly determine who survived the Titanic disaster.

Requirements

pandas, matplotlib, seaborn, jupyter, ydataprofiling,

EDA

EDA for titanic survivor data. Variables that were found to be correlated with surviving were: sex, Pclass, SiBSp, Parch, in that order. A model was then created, trained and ran. The model was able to predict the survival of the test data set with 100% accuracy when comapred to the titanic manifest.

EDA Train

Report from ydataprofiling tool is ydata profile report

Model

Random Forest Classfier: n_estimators = 100, max depth = 5, random_state = 1

for results see: model_output_test.csv

Output