AgePredictor VS Smile

Even funnier: perhaps we should run a linear-interpolation or entropy classifier on our respective texts and get an age estimate. I'm betting it won't reflect on you well.

> I don't think it's right to recommend that new users move away from the package because of licensing issues

I was going to chime in to agree but then I saw how this was done - a completely innocuous looking commit:

https://github.com/haifengl/smile/commit/6f22097b233a3436519...

And literally no mention in the release notes:

https://github.com/haifengl/smile/releases/tag/v3.0.0

I think if you are going to change license especially in a way that makes it less permissive you need to be super open and clear about both the fact you are doing it and your reasons for that. This is done so silently as to look like it is intentionally trying to mislead and trick people.

So maybe I wouldn't say to move away because of the specific license, but it's legitimate to avoid something when it's so clearly driven by a single entity and that entity acts in a way that isn't trustworthy.

Check out Smile too.

Although it may be a bit more work, you can do both machine learning and AI in Java. If you are doing deep learning, you can use DeepJavaLibrary (I do work on this one at Amazon). If you are looking for other ML algorithms, I have seen Smile, Tribuo, or some around Spark.

For deploying a trained model there are a bunch of options that use Java on top of some native runtime like TF-Java (which I co-lead), ONNX Runtime, pytorch has inference for TorchScript models. Training deep learning models is harder, though you can do it for some of them in DJL. Training more standard ML models is much simpler, either via Tribuo, or using things like LibSVM & XGBoost directly, or other libraries like SMILE or WEKA.

I use smile https://github.com/haifengl/smile with ammonite and it feels pretty easy/good to work with. Of course for pure looking at data, and exploration, you're not going to beat python.

Actually, it does. Scala has Spark for data science and some ML libs like Smile.

I guess I must have a mistake somewhere. It's not much code. it's written in Kotlin with smile. My dataset is only about 32MB. I load the dataset into memory. I then use 80% of the data for training, and the other for later testing. I get just the columns I need and store them in the variable dataset.

I've heard good things about Smile, probably beats libs like Weka by far. I'm not sure if you can load a scikit-learn model though, so you might need to retrain the model in Kotlin.

I was using Smile for some period - https://haifengl.github.io/ - it's quite small and lightweight Java lib with some very basic algorithms - I was using in particularly cauterization. Along with this it provides Scala API.