[tcrews]

In the name of reproducible scientific results, is the training database available somewhere?

[rightsForRobots]

The photos aren't available but the trained model is. It's a set of weights.

[JulianMorrison]

Run it backwards in inception mode? I wonder if it could generate porn, or what Gigeresque horrors would come out if somebody tried.

[cing]

Image Synthesis from Yahoo's open_nsfw (https://news.ycombinator.com/item?id=12756462)

[rightsForRobots]

I actually tried that. It's not as interesting as you'd think. Perhaps having an 'other' category makes it more difficult.

It's trivial to drop the model into the deepdream ipython notebook they provide: https://github.com/google/deepdream/blob/master/dream.ipynb

[rightsForRobots]

Tried again. Better: http://i.imgur.com/ALoLmcX.jpg

[theoh]

Seriously, though, synthesis using a recognition model can be a good reality check to remind us of the shortcomings of the model's "understanding" of the domain.

[SparkyMcUnicorn]

It looks like it's the 22MB file in the model folder.

https://github.com/ryanjay0/miles-deep/tree/master/model

[chrisfosterelli]

This is just the resulting network weights. While useful, I believe this alone wouldn't let someone reproduce the experiment as easily as having the dataset.

[rightsForRobots]

I don't have the rights to the dataset, so unfortunately I won't be releasing it.

[dharma1]

You should be able to scrape 30k+ labeled images relatively easily

[suryabhupa]

How do you imagine going about this?

[dharma1]

in the case of porn, there is a zillion sites of images with labeled categories. Web scrape those