Alzheimers Classification using Brain MRI's

Usage

Dependencies

pip install -r requirements.txt

Training

Navigate to src/

python train.py

Parameters:
--epochs number of epochs
--lr learning rate
--batch batch size
--model model_name
--num_classes number of classes
--gpu to use gpu
--optimizer type of optimizer, default:adamax --logger tensorboard(tb) or wandb(eb)

Dataset

Download the dataset from the link below to 'data/' directory Kaggle : Alzheimers Brain MRI

Description

-   Number of Classes : 3
-   Image Size: 2048 x 2048
-   Total Images : 600
    -   AD : 200
    -   MCI : 200
    -   CN : 200

Models

• EfficientNet
• ResNet50
• ViT
• ConvNeXt