Skip to content

CNN Network based on VGG16 for classification of 10 types of animal images.

License

Notifications You must be signed in to change notification settings

marcus-deans/Animal10-VGG-Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Animal10-VGG-Classification

CNN Network based on VGG16 for classification of 10 types of animal images.

I developed a convolutional neural network model for recognising animals from the Kaggle Animal10 database, consisting of approximately 28k medium-quality images. Transfer learning is implemented using VGG16 pre-trained weights for higher efficiency, with a final classification model constructed around this. Hence, this CNN uses 13 convolutional layers, 5 pooling layers, and 2 fully connected layers to classify the images in Animal10 as the appropriate animal. This model achieves 95% test accuracy within 12 epochs, with runtime of approximately 30 minutes using Kaggle GPUs.

The model was built in Kaggle for easy dataset import and operations, and the original script may be viewed here, along with the execution output.

Libraries/frameworks implemented within this project:

  • Keras
  • TensorFlow
  • NumPy
  • pandas
  • matplotlib
  • scikit-learn

About

CNN Network based on VGG16 for classification of 10 types of animal images.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages