On: Show HN: Basilica – word2vec for anything

[asdfghjl]

How are you embedding images?

[mlucy → asdfghjl]

We're feeding them through a deep neural network and using the activations of an intermediate layer as an embedding.

You can read more about this technique in https://arxiv.org/abs/1403.6382 if you're interested.

[PeterisP → asdfghjl]

For most purposes, taking a decent ImageNet model and ripping off a couple final layers works reasonably well.