Tuesday, August 2, 2016

Caffe Deep Learning iPython Notebook

Added an ipython notebook at Git: https://github.com/singarajus/caffepythonnotebook

It is based on Caffe's python bindings for classify.py.

In the notebook, set a url for an image, hit the button: "Submit for run:". All the following cells will be computed. The notebook will identify, "Convolution", "InnerProduct" layers and create drop down boxes. Selecting any convolution layer will allow, visualization of that layer.

The model used in the notebook was imagenet.

A sample input file:

Filters at convolution layer 5


Layer activation at Convolution Layer 5:

Output Probability: