CaffeJS - User Guide

To run the samples, please execute npm start from the root directory. Make sure you updated all dependencies by running npm install beforehand.

Running CaffeJS

Clone the repository to your local drive and then start a static webserver in the root directory (e.g. run npm start). Now you can open models.html that loads and visualizes popular deep learning architectures (such as AlexNet, VGG, GoogLeNet, etc.) to analyze their structure. This loads and parses the Caffe models from .prototxt files and convert them on-the-fly to ConvNetJS models.

To run a forward pass we need to load some pretrained model weights. First make sure you download the model weights for your particular model by running sh fetch_weights.sh in the model directories. This will pull and unzip the binary model weights optimized for JavaScript form my Dropbox. These weights have been generated via the scripts/convert_caffemodel.py.

Loading the labels

Ỳou can load the ImageNet labels from the data directory, using the following snippet.

var labels;
d3.text('data/ilsvrc12/synset_words.txt', function(data){
  labels = data.split('\n').map(function(d){
    return d.substr(10);
  });
});

Using the mean.binaryproto values

To use the mean.binaryproto values in CaffeJS, we need to parse the binary files into Volumes. First, we need to include to Protobuf specific libraries in the header.

<script src="libs/long.min.js"></script>
<script src="libs/bytebuffer.min.js"></script>
<script src="libs/protobuf.min.js"></script>

Now we can use the BlobProtoParser to transform the binary file into a Volume.

var mean;
var p = new Parser.BlobProtoParser();
p.parse('models/age_net/mean.binaryproto').then(function(data){
  mean = data;
});

You can debug the mean values by rendering the Volume to the screen. Add the following line to the callback function to render the Volume to a Canvas.

ImgJS.Image.fromMean(mean).render();