Skip to content
A SVM that can predict the weather using data from https://catalog.data.gov/dataset/local-weather-archive. Received third place at the McMaster Engineering Competition 2017 (Programming Division).
Java Python
Branch: master
Clone or download

Latest commit

Fetching latest commit…
Cannot retrieve the latest commit at this time.

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Documentation_Assets
build
nbproject
src/examples
.gitignore
A Weather Prediction System Report.pdf
README.md
SVM.py
Weather Prediction System Powerpoint.pdf
build.xml
manifest.mf
weather_data.csv

README.md

McMaster Engineering Competition (Programming Division) 2017 Weather Prediction Application - Team Kelvin and Friends

UPDATED: DECEMBER 24, 2017


NOTE: This project is currently undergoing revisions to improve its performance and reliability. Part of this includes updating the method used to process the data to allow for easier interpretation, and testing the functionality of the checkboxes. Plans are also in place to bring this project to a more presentable state than what was originally built during the day long competition.


This project predicts weather using data from the Raleigh-Durham International Airport collected by NOAA. It was created for the McMaster Engineering Competition (Programming Division) 2017, where it received third place out of approximately 20 participating teams.

Contents of the Repository

  1. The source code for the GUI is located in src/examples.
  2. The source code for the SVM is located in the root under SVM.py.
  3. A Makefile is located in src
  4. The presentation used in the competition is located in the root directory.
  5. A design document is available in the root directory.
  6. All auxiliary files used for the presentation and design document are located in the Documentation_Assets folder.

Using the Program

  1. Under the "Weather" section of the application, select the factors you want to consider in the application.
  2. Under the test section, enter artificial conditions to reflect real world sensor data.
  3. Press submit to generate a prediction.
You can’t perform that action at this time.