VGG16 in Keras

You can follow along with the code in the Jupyter notebook ch-12a_VGG16_Keras.

Now let us do the same classification and retraining with Keras. You will see how easily we can use the VGG16 pre-trained model in Keras with the lesser amount of code.

Image classification using pre-trained VGG16 in Keras

Loading the model is a one-line operation:

from keras.applications import VGG16
model=VGG16(weights='imagenet')

We can use this model to predict the probabilities of classes:

probs = model.predict(images_test)

Here are the results of this classification:

Input image	Output probabilities
	Probability 99.41% of [zebra] Probability 0.19% of [tiger cat] Probability 0.13% of [goose] Probability 0.09...