Dog Breed Classifier

Upload a photo of a dog, and our machine learning model will identify its breed. This model can recognize over 120 different dog breeds with high accuracy.

Drag and drop an image here, or click to select

Supported formats: JPG, JPEG, PNG

About This Project

This dog breed classification system uses a deep convolutional neural network (CNN) trained on thousands of dog images. The model has been fine-tuned to recognize subtle differences between various dog breeds.

The model architecture is based on a pre-trained ResNet50 backbone with custom classification layers added on top and fine-tuned specifically for dog breed recognition.

The API endpoint is hosted on Render, and this web interface communicates with it to get predictions. The model can identify over 120 different dog breeds with an accuracy of approximately 90% on validation data.

Technologies Used:

  • Python
  • PyTorch / TensorFlow
  • FastAPI
  • Flask
  • HTML/CSS/JavaScript