Ballot Paper Classification - Computer Vision Project

Demo

Model

The model is trained on ResNet50 architecture. Model and encoder can be downloaded from the following link:

https://www.floydhub.com/cruzaan/datasets/ballot-model/1/

Store the modeland encoder in artifacts directory.

Prepare the data for Evaluation

Prepare your dataset to proceed with the model. Your data folder structure should be on following format:

│
├── data             <- Root data folder
|   |── 0             <- Sub data folder with it's associated label
|       |── img1.jpeg   <- Image file with same filename as on csv file
|   |── 1
│

To structure data this way, while running evaluate.py make sure to change process_data param to True

Evaluate data

The evaluate.py function will evaluate the provided data and returns the evaluation report as an output.

python evaluate.py

Flask App

To run the flask app, use the following command.

python app.py