https://github.com/alexweininger/aurora-classifier
Perceptron machine learning algorithm that classifies images as either containing aurora or not containing aurora.
https://github.com/alexweininger/aurora-classifier
Last synced: 7 months ago
JSON representation
Perceptron machine learning algorithm that classifies images as either containing aurora or not containing aurora.
- Host: GitHub
- URL: https://github.com/alexweininger/aurora-classifier
- Owner: alexweininger
- Created: 2019-02-10T06:39:57.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2019-02-10T06:49:50.000Z (almost 7 years ago)
- Last Synced: 2024-10-18T06:16:45.191Z (about 1 year ago)
- Language: Python
- Size: 1020 KB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# How to use
## Dependencies
- Python 3
- pandas
- matplotlib
- csv
## histograms.py
There are two programs involved with this application. The first `histgrams.py` generates the histograms from the images and outputs a text file containing the histograms on each line. You do not need to run `histgrams.py` unless you want to change the images tested. The lines are prepended with whether or not the image contains an aurora (for validation and testing). The `output_stretch.txt` file is already generated and you do not need to run `histgrams.py` again. If you want to change the images tested, put them into the respective folders in the `training_images` directory.
## perceptron.py
This is the main program of this application. To run simply type `py perceptron.py`. The final accuracy is printed at the end of the programs execution. This test is run off different images than the training data.