I am doing PCA with sklearn library in python. I have seen a lot of questions on StackOverflow about this attribute, but I still not understand what are the 'loadings' typed on these questions. I have a data composed as (x_sample, y_features) (100, 1577)
, when I compute the PCA on this data it becomes (100,90) (x_sample, y_components (the features reduced))
. After I compute the attribute pca.components_
, it becomes (90,1577)
so in a shape (number_components, original_features)
. What can I say about this attribute? My goal is to extract some relevant features in unsupervised learning and I want to know if this attribute will be useful. Thank you for your help.
question from:
https://stackoverflow.com/questions/65905082/what-is-the-pca-components-attribute-in-sklearn 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…