Skip to content

Repository files navigation

Description

In this project, I conducted a comprehensive comparison between a scratch CNN model and fine-tuned models, evaluating their performance on the Intel Image Classification dataset. These approaches were analyzed in terms of accuracy, training efficiency, and suitability for the task, providing valuable insights into model selection and optimization strategies.

Commands

pip install -r requirements.txt

About Dataset

Size : 363 MBs
Link : https://www.kaggle.com/datasets/puneet6060/intel-image-classification
Classes: Buildings, Forest, Glacier, Mountain, Sea, Street
Folders: Train, Test and Prediction
Total Images: 24,000+

About Models

A. Built From Scratch

Convolution Neural Network - IntelNet

B. Comparitive Analysis

  1. Finetuned: MobileNet-V2
  2. Finetuned: Wide ResNet-50
  3. Finetuned: VGG-16

About

Classifying image data of Natural Scenes around the world.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages