Code to create montages / question about backpropegation in max tracker

[cfunk1210]

I can't find the code to create the montage images from the cropped activations and deconvolutions. Is it included or could you release the code for that?

Also, at line 309 of max tracker, you set the diff to 1.0 but doesn't it make more sense to set it to the recorded_val so that it is deconvoluting/backpropegation based on the maximum strength of the activation? This also differs from the tool since that backpropagate the entire activation layer instead at line 89 of caffe_proc_thread.py diffs[0][backprop_unit] = self.net.blobs[backprop_layer].data[0,backprop_unit]

[arikpoz]

@cfunk1210 I've substantially refactored the tool, made the whole image generation outputs structure simpler and loadable into the tool without further post-processing

Latest version is in https://github.com/arikpoz/deep-visualization-toolbox

[Huxwell]

@arikpoz Could you please point where in your version of the tool we can find the montages code?

[arikpoz]

@Huxwell I don't have montages code, instead I made the tool also able to load the output of the offline scripts without further processing (since the montage code is missing)