## Introduction
This program is a python project that will demonstrate the use of Convolutional Neural Networks (CNNs) on the Ugandan Passion Fruit Disease Detection Prediction Challenge provided by Zindi.
The program uses Keras from Tensorflow in order to create, train and test the CNN.
## Links
* [Zindi](https://zindi.africa)
* [Passion Fruit Disease Detection Dataset](https://zindi.africa/competitions/makerere-passion-fruit-disease-detection-challenge/data)
* [Keras](https://keras.io/)
* [CNNs](https://en.wikipedia.org/wiki/Convolutional_neural_network)
## Setup
1. Clone the repo into the desired folder
2. The following directory structure was used. If there is a need to replicate, please use the same structure.
If the data files are placed in a different directory, the base parameter may be set in main.py to point to it.
+-- Data | +-- Train.csv | +-- Test.csv +-- Networks | +-- CNN.py +-- Test_Images +-- Train_Images +-- Data.py +-- main.py +-- Network.py +-- requirements.txt
3. Run the program by typing in console
python main.py
4. You may need to install modules, this can be done by typing the below command - All required modules are stored in the requirements.txt file
pip install -r requirements.txt
## General Info
This repo is an Honours level assignment for the University of Pretoria South Africa (COS 711). \
Please feel free to email me if there's any questions, my email can be found on my profile.
The structure of the project allows for additional Networks to be created and added.