Dilated Residual Networks (Custom Dataset for Image Classification task)

Author

Arpit Aggarwal

Introduction to the Project

In this project, different CNN Architectures like DRN-34, DRN-50, ResNet-34 and ResNet-50 were used for the task of Dog-Cat image classification. The input to the CNN networks was a (224 x 224 x 3) image and the number of classes were 2, where '0' was for a cat and '1' was for a dog. The CNN architectures were implemented in PyTorch and the loss function was Cross Entropy Loss. The hyperparameters to be tuned were: Number of epochs(e), Learning Rate(lr), momentum(m), weight decay(wd) and batch size(bs).

Data

The data for the task of Dog-Cat image classification can be downloaded from: https://drive.google.com/drive/folders/1EdVqRCT1NSYT6Ge-SvAIu7R5i9Og2tiO?usp=sharing. The dataset has been divided into three sets: Training data, Validation data and Testing data. The analysis of different CNN architectures for Dog-Cat image classification was done on comparing the Training Accuracy and Validation Accuracy values.

Results

The results after using different CNN architectures are given below:

1. ResNet-50(pre-trained on ImageNet dataset)
   

Training Accuracy = 99.43% and Validation Accuracy = 98.43% (e = 50, lr = 0.005, m = 0.9, bs = 32, wd = 5e-4)

2. DRN-50
   

Training Accuracy = 96.91% and Validation Accuracy = 94.53% (e = 100, lr = 0.005, m = 0.9, bs = 32, wd = 5e-4)

3. DRN-34
   

Training Accuracy = 98.23% and Validation Accuracy = 96.68% (e = 100, lr = 0.005, m = 0.9, bs = 32, wd = 5e-4)

Software Required

To run the jupyter notebooks, use Python 3. Standard libraries like Numpy and PyTorch are used.

Credits

The following links were helpful for this project:

1. https://github.com/fyu/drn
2. https://towardsdatascience.com/review-drn-dilated-residual-networks-image-classification-semantic-segmentation-d527e1a8fb5