No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
submission.csv
test.csv
titanic.py
train.csv

README.md

kaggle-titanic

About

This repository contains the code that achieves 76% score in the Kaggle Titanic Competition The code was developed to serve as an initial exploration of Scikit-Learn.

Pre-Processing

To achieve the result, the following were the pre-processing techniques applied:

  • [Age], [Fare]: Missing Values were handled with Mean
  • [Embarked]: Missing values were handled with Most Frequent Strategy
  • [Embarked]: One Hot Encoder applied
  • [Sex]: Ordinal Encoder applied
  • [Age], [SibSp], [Fare], [Pclass]: Numerical Features that were normalized with MinMaxScaler

Model

The model built in this implementation is a simple Logistic Regression Model