https://github.com/alishobeiri/artificial-stupidity
Kaggle competition project using a modified version of the Google Quick, Draw dataset
https://github.com/alishobeiri/artificial-stupidity
Last synced: 3 months ago
JSON representation
Kaggle competition project using a modified version of the Google Quick, Draw dataset
- Host: GitHub
- URL: https://github.com/alishobeiri/artificial-stupidity
- Owner: alishobeiri
- Created: 2018-12-15T15:07:24.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2018-12-17T14:17:12.000Z (almost 7 years ago)
- Last Synced: 2025-02-14T13:50:30.347Z (8 months ago)
- Language: Jupyter Notebook
- Homepage: https://www.kaggle.com/c/f2018-hand-drawn-pictures
- Size: 58.6 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Artificial-Stupidity
Elie-Joe Haykal, Antonios Valkanas, Ali Shobeiri
This repo contains our team submission to: https://www.kaggle.com/c/f2018-hand-drawn-pictures
#### Technologies used: *SciKit Learn, OpenCV, Keras*
OpenCV was used primarily to reduce image noise
SciKit Learn was used to fit linear baselines to measure performance
Keras was used to apply Deep Learning to the denoised images for classification
#### File Breakdown
[KaggleProject.ipynb](https://github.com/alishobeiri/Artificial-Stupidity/blob/master/KaggleProject.ipynb): Contains our best performing deep learning model as well data preprocessing steps to reduce image noise.
[Neural-Network.ipynb](https://github.com/alishobeiri/Artificial-Stupidity/blob/master/Neural-Network.ipynb): Contains a hand implemented fully connected feed forward neural network that is used to classify images.
[Linear-SVM.ipynb](https://github.com/alishobeiri/Artificial-Stupidity/blob/master/linear_SVM.ipynb): Contains benchmarking used to determine performance of preprocessing and give baselines to compare our deep learning solutions to.