This repo contains the files and data for a CNN that predicts the health of leaves. The model was created for Hack Western 2018, the goal of the model was to determine the health of a crop based on a photo of its leaves. Different illnesses can manifest ad spots and discoloration on the leaves of crops and a model can be trained to recognize these characteristics based on photos of healthy and unhealthy leaves.
Future updates:
- allow model variables to read from a file instead of hardcoded
Leaf data is complied from https://github.com/salathegroup/plantvillage_deeplearning_paper_dataset/tree/master/raw The tensorflow model was created with the help from https://pythonprogramming.net/loading-custom-data-deep-learning-python-tensorflow-keras/?completed=/introduction-deep-learning-python-tensorflow-keras/ tutorial