This library differs in that it implements its own matrix operations and does not rely on Gonum. It is really a collection of implementations that include the following:

• Linear regression
• Logistic regression
• Neural networks
• Collaborative filtering
• Gaussian multivariate distribution for anomaly detection systems

Individually, these are powerful tools; indeed, linear regression is often described as one of the most important tools in the data scientist's toolkit, but, for our purposes, we really only care about the neural networks portion of the library. And here, we see limitations similar to those of GoLearn, such as limited activation functions and a lack of tools for intra- and interlayer connections (for example, LSTM units).

The author has an additional library that implements CUDA matrix operations; however, both this and the go_ml library itself have not been updated in four years (at the time of writing), so this is not a project you could simply import and start building neural networks straightaway.