The original code could only keras version 2.0.9 under.
So, I modified little line. Usage is same like original version.
$ python grad-cam.py <path_to_image>
below script is from original github(https://github.com/jacobgil/keras-grad-cam)
Gradient class activation maps are a visualization technique for deep learning networks.
See the paper: https://arxiv.org/pdf/1610.02391v1.pdf
The paper authors torch implementation: https://github.com/ramprs/grad-cam
This code assumes Tensorflow dimension ordering, and uses the VGG16 network in keras.applications by default (the network weights will be downloaded on first use).
Usage: python grad-cam.py <path_to_image>
Example image from the original implementation:
'boxer' (243 or 242 in keras)
'tiger cat' (283 or 282 in keras)