A simple Cosine similarity
- Vectors contains one column, m rows : each row contains an array of n floats;
- I want to have a m x n matrix, transpose it and do there product
- I’m using C#
Also, I read this from official documentation https://numerics.mathdotnet.com/Matrix.html
"For example, if you have a collection of vectors, consider to store them in a list or array of vectors, not in a matrix (unless you need matrix operations, of course)."
So please what’s the optimized way to achieve this with Math.net.numerics? Thanks in advance
#UPDATE
I’m getting this Error
CS1503 Argument 3: cannot convert from'System.Collections.Generic.IEnumerable<float[]>' to 'float[]' With this code :
IEnumerable<float[]> Vectors = predictions.GetColumn<float[]>("Features"); var x = new M.DenseMatrix(1, Vectors.Count(), Vectors); 
Sample method to do the cross-product using MathNet.Numerics.LinearAlgebra.Generic.Vector of a 2 element vector.
Vector<float> v1 = new DenseVector(new float[] { 1, 2, 3 }); Vector<float> v2 = new DenseVector(new float[] { 3, 2, 1 }); Vector<float> result = v1.CrossProduct(v2); Vector<float> result = Vector.CrossProduct(v1,v2); 
