Plant Village Classification
This project aims to use different tools available in the field of machine learning and neural network training to build and deploy a model to classify plant diseases. The task and the dataset itself is not the focus of this project. The particular dataset was chosen because:
- It is a well-known dataset in the computer vision community.
- It is a fairly large dataset, which allows us to train a reasonably good model.
- The dataset is simple enough to be used as a tutorial for building and deploying a machine learning model.
Goals of this project
The main goals of this project are:
- Familiarize with data version control using DVC.
- Build a neural network model using PyTorch to classify plant diseases.
- Track experiements using Weights & Biases.
- Use ONNX to export the trained model and deploy it using Flask
Getting Started
Preparing a repository
This entire project will be built in a Git repository. Create an empty repository on GitHub. Don't forget to initialize .gitignore for python files. Now clone the repository to your local machine:
Setting up the environment
Here, we use conda to manage our environment. You can use any other tool you are comfortable with. Create a new conda environment and activate it:
Let's install some of the dependencies we will need for this project: