/lighthouse-mid-term-project

Lighthouse Labs Data Science Midterm Project

Primary LanguageJupyter Notebook

lighthouse-mid-term-project

Lighthouse Labs Data Science Midterm Project

Slide Deck https://slides.com/gailbishop/midterm-project

Table of Contents

  1. db_connection - this notebook creates the Postgres db connection and is imported into other workbooks that require the connection
  2. Flights Dataframe - this workbook has exploratory data analysis of the Flights data
  3. Flights Dataframe 2 - this workbook has exploratory data analysis of the Flights data and related Fuel Consumption and Passengers data
  4. Feature Creation Methods - this workbook contains all methods to create the new features and stores the dataframe as a compressed csv
  5. Reference Dataframes - this workbook contains methods to create summary reference tables of data for creating aggregate features
  6. WeatherData - this workbook determines a list of 52 representative cities that were used to query the Weather API, and queries the Weather API for 12 months of aggregate weather data
  7. Regression Model Evaluation - this workbook run 3 regression models and shows evaluation methods for each
  8. Run Final Model - this workbook read the Test Data from a csv, runs the saved XGB model and exports the predictions to a csv
  9. submission_xgb.csv - the prediction file for the first week of January 2020

Reference Folder

This contains Pickle files of dataframes created by the Reference Dataframe workbook. These pickle files were then imported into the Feature Creation Methods workbook.