A PyTorch-based musical mel-frequency transform for audio processing, optimized for performance and ONNX compatibility.
If you've ever wanted audio features that directly represent semitones (or quarter tones!) this is the package for you.
132 karma · joined December 6, 2013
If you've ever wanted audio features that directly represent semitones (or quarter tones!) this is the package for you.
The Ray ecosystem is actually chalk full of integrations, from XGBoost Ray (https://docs.ray.io/en/master/xgboost-ray.html), to PyTorch on Ray (https://docs.ray.io/en/master/using-ray-with-pytorch.html), and of course hyperparameter search with Ray Tune for a variety of libraries, including Sklearn (https://github.com/ray-project/tune-sklearn).
I’ve found that opacity if 0.7 is often a nice compromise.
Basically what this will do is assign each tile less than 100% of the time randomly and then for each that isn’t, you choose among the top best K matches on L2 distance with equal probability. Gives it a little bit of both.
L*b was very slow, yes. It’s all done offline but I tend to like quicker feedback...
As to why it was in the project: If you treat each pixel as an example vector in 3 dimensions and cluster, you get the “dominant” colors for the image. It’s a primitive way to compress images as well. In this case I just was using it to generat fun cards that would use a minimal number of dominant colors. It’s still in the code if you’d like to use it but a bit hidden.
But you’re correct, generally that’s the space you want to be in.
As I mentioned above the only ML is the face classifier. There’s a flag that allows you to only include face pics in the photomosaics as well
But yes the photomosiacs strictly don’t use ML, unless you count the internal fun stuff Faiss (the similarity search lib) does to construct fast indexes.
Do you have planned support for consensus-type mutlisig access? (ie: needing M of N approvals to acccess resources)
80% of the problems Coinbase and every other exchange has are dealing with "Finance 1.0" banks & payment processors who still move funds around by having their employees manually type in numbers and sending transfer records via FTP. Seriously.
In a sense, Coinbase and other exchanges are moving money out of the traditional (fragmented) banking system and into a digital one. That takes friction and dealing with fragmented systems.
And personally I'd trust Coinbase over any of the other ones any day of the week. Have you dealt with Kraken/Poloniex/etc? Terrifying.
This isn't really news.
The entire point of using stochasticity (ie: random shuffling) is to avoid similar and/or a same-ordered run of examples from redirecting the hill climbing in a globally non-optimal direction all at once.
A single user's examples will be very similar, so you can shuffle all the examples from one user you want - that doesn't make it truly stochastic in the context of gradient descent optimization.
The quantization / compression part is pretty cool though. I suppose that could obfuscate slightly what the original example was for privacy purposes? Seems like you'd lose on accuracy though.
It's just a relative measure of confidence, scaled such that they all sum to 1.0.
Most people don't realize this, but usually artists/labels will have a different mastered version of a song on each platform. Spotify, for instance, has it's own normalization algorithm that it puts tracks through to even out the listening experience in terms of loudness (RMS). Of course artists and their mastering engineers want to have some control over how that sounds, so they will change it.
You could, for instance, insert other tracks into the audio additively to try and confuse the fingerprint retrieval logic into suspecting a different track, but since this and many other fingerprinting techniques depend on the actual frequency of the audio emitted, there's no shortcut to obscuring the actual track.
The American vernacular is you get "dejavu" whenever you have that "I've been here before..." feeling.