mrinaljain17 / vgg

Keras implementation of VGG from the paper "Very Deep Convolutional Networks for Large-Scale Image Recognition" - https://arxiv.org/abs/1409.1556

VGG

Keras implementation of VGG from the paper Very Deep Convolutional Networks for Large-Scale Image Recognition

Insturctions

1. Clone the repository and navigate to the downloaded folder.

	git clone https://github.com/MrinalJain17/VGG.git
	cd VGG

2. Import the module and get the required model.

	from vgg import VGG
	
	# Getting VGG-16 (16-layer VGG model) for a dataset with 20 classes
	model = VGG(model_type='D', num_classes=20)
	model.summary()

3. The model was tested on the cifar-10 dataset.
   For further details, view the jupyter notebook by running the command:

	jupyter notebook cifar-10.py

Requirements

Python 3.x (preferably from the Anaconda Distribution)

Additional Libraries to install

• Keras

	pip install keras

(With backend as Tensorflow)
For installation of Tensorflow, view instructions on their site here.