At https://mccluskey.scot/trad_ml_methods/pca.html
We can visiualise these as vectors over our data, where the components are the vector direction and the explained variance is the squared-length.
These vectors are the data’s principal axes and then length is
(it is a measure of the vairance when the data is projected onto that axis)
arm61
commented
3 years ago
At https://mccluskey.scot/trad_ml_methods/pca.html We can visiualise these as vectors over our data, where the components are the vector direction and the explained variance is the squared-length.
These vectors are the data’s principal axes and then length is
(it is a measure of the vairance when the data is projected onto that axis)