Create model which when give an image of a crop infested with disease its able to detect which type of disease
Download Anaconda for your the specific platform
Create conda environment for Python3
conda create -n "environment_name" python=3.7
Activate the environment.
conda activate environment_name
Clone the repo
https://github.com/r-wambui/Agro-detect-model.git
Navigate to the root folder
cd Agro-detect-model
create a data directory
mkdir data
-
Download the dataset here. I'm using the dataset without augmentation and uzip the files.
-
Rename the folder to
dataset. -
We need to split this dataset into train and test. On the terminal run
python split_data.py
-
This will create two folders train and val.
-
Run
jupyter notebookthen open Agro-detect.ipynb -
You can now interact with the model and follow the tutorial Build a Simple Crop Disease Detection Model with PyTorch.