benw413 / data_science_general_assembly Goto Github PK

Hello World! This repository contains multiple projects from my Data Science Immersive course at General Assembly. The first project I have posted in this repository is a series of three separate Jupyter notebooks under the Titanic_Series_Commit folder. This series contains a progression from basic data analysis workflow to professional level workflow using machine learning in conjunction with the Titanic dataset from Kaggle.com. The purpose for these notebooks are to serve as a template/tutorial for how to progress from the basics of descriptive analytics to the more advanced techniques of machine learning. Over three separate Jupyter notebooks, I cover a step by step breakdown of how to explore, clean, analyze, visualize, and predict survival for Titanic passengers. I will demonstrate everything from importing libraries to uploading your predictions to Kaggle in a format that it will accept. Finally, I will also demonstrate how to automate workflow using pipelines to make code re-useable.

If you would like to learn more info about my Data Science Experience please feel free to check out my Blog at https://medium.com/@benweinstein_52172

This notebook uses pandas to explore, clean, analyze, and visualize basic information about the Titanic dataset and includes findings related to survival based on age, gender, class, and a host of other characteristics.

This is a step by step guide to cleaning a dataset, manipulating the features, creating a logistic regression model, obtaining predictions, and passing the dataset into a pandas dataframe to upload to Kaggle. As you will see, my process is line by line to help with visualization however not extremely reusable. I will explain how to make my code reusable in the next series with pipelines.

In this third and final jupyter notebook I will demonstrate how to use pipelines to allow for code to be re-useable. I will also upload this final dataset to Kaggle.com and report back a score.

This project is a commit from the Kaggle Housing Prices competition. It is currently under contruction so check back in later for an update.