Machine Learning in Go

WORK IN PROGRESS

This is an attempt to learn machine learning and the necessary mathematics by way of coding.

The plan is to build a simple machine learning library using only standard libraries and refine it as more complex usage arises.

Functionalities

Math related:

• Handling ONLY 2D matrixes (for now)
• Matrix transposition
• Simple matrix arithmetics, scalar-matrix arithmetics and matrix dot product
• Matrix initializations with random numbers and zeros
• Retrieval of highest and lowest value inside a matrix
• Matrix shuffling
• Get matrix size and shape (as either integers or formatted string)

ANN related:

• Handling simple neuron layers consisting of weights and biases
• Initialize a simple ANN consisting of one input, hidden, and output layer
• Do Gradient Descent iterative optimization algorithm on a dataset

Data encoding related:

• Do one-hot encoding on the data's labels