https://github.com/mohaimenulislamshawon/cnn-image-classification-cheetah-vs-hyena

The project demonstrates a CNN image classifier model "cheetah vs hyena predictor" deployed via a web interface.
https://github.com/mohaimenulislamshawon/cnn-image-classification-cheetah-vs-hyena

binary-classification cheetah-hyena-predictor cnn convolutional-neural-networks image-classification machine-learning-model-deployment neural-network webview

Last synced: about 1 month ago
JSON representation

The project demonstrates a CNN image classifier model "cheetah vs hyena predictor" deployed via a web interface.

• Host: GitHub
• URL: https://github.com/mohaimenulislamshawon/cnn-image-classification-cheetah-vs-hyena
• Owner: mohaimenulislamshawon
• License: apache-2.0
• Created: 2023-12-10T18:03:03.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-01-28T09:03:06.000Z (12 months ago)
• Last Synced: 2024-01-28T12:22:25.715Z (12 months ago)
• Topics: binary-classification, cheetah-hyena-predictor, cnn, convolutional-neural-networks, image-classification, machine-learning-model-deployment, neural-network, webview
• Language: Python
• Homepage: http://cheetah-hyena-predictor.mislamshawon.repl.co/
• Size: 14.6 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# CNN Image Classification: Cheetah vs Hyena

The project demonstrates a CNN image classifier model "cheetah vs hyena predictor" deployed via a web interface.

This project involved developing a convolutional neural network (CNN) model to classify images as either cheetahs or hyenas using a Kaggle dataset. Data preprocessing steps included resizing images to 200x200 pixels and normalizing pixel values to the 0-1 range. 

The CNN model architecture consisted of multiple convolutional and max pooling layers, followed by fully connected layers and a final sigmoid output layer to make a binary prediction. Model training was done for 100 epochs with early stopping to prevent overfitting, achieving over 94% validation accuracy. 

Key evaluation metrics on the test set were 50% accuracy, 50% precision, 52% recall and 51% F1-score. For model serving, a Flask web application was built to allow image uploads and display the cheetah vs hyena classification result in the browser. 

Overall, the project demonstrates an end-to-end implementation of a CNN image classifier deployed via a web interface.

# Download the model & add to directory 

https://drive.google.com/file/d/1AI07wK1d22hvBFNcWOse2BHO6lwnV21-/view?usp=sharing

# Dataset

https://www.kaggle.com/datasets/singhdatascientist/for-image-classification-of-cheetah-vs-hyena

# Notebook that I have Created

https://www.kaggle.com/code/mohaimenulshawon/cnn-image-classification-cheetah-vs-hyena

# Web interface that I have Created

http://cheetah-hyena-predictor.mislamshawon.repl.co/