Python Deep Learning Cookbook

How to do it...

1. We start by implementing all Keras tools:

from keras.models import Modelfrom keras.applications.resnet50 import ResNet50from keras.applications.resnet50 import preprocess_inputfrom keras.preprocessing.image import load_imgfrom keras.preprocessing.image import img_to_arrayfrom keras.applications import imagenet_utils

2. Next, we load the ResNet50 model with the imagenet weights:

resnet_model = ResNet50(weights='imagenet')

3. For this example, we will extract the final average pooling layer in the ResNet implementation; this layer is called avg_pool. You can look up the Keras implementation of ResNet50 at https://github.com/fchollet/keras/blob/master/keras/applications/resnet50.py if you want to extract the features from another ...