Malware-Classification-using-Malimg-dataset

Problem Defination:

Malware is any malicious code or a program that can be harmful to the computer. 
There are many types of malwares, and it’s essential to detect these types to prevent their breaches to keep the data and the system private and secured.

Dataset:

- The Malimg dataset consists of 9339 images and 25 classes. 
- The dataset contains the family/class of the malware and the malware type.
- The data has been split into training, testing and validation sets.

Conclusion:

- After training AlexNet, VGG16 and ResNet101, the champion model was ResNet101 (after handling imbalanced data) with test accuracy of 97.7%.