70 likes | 103 Views
Dive into the world of PCA, a statistical technique invented in 1901, used for data reduction, compression, computer vision, face recognition, and pattern identification. Learn the six essential steps for implementing PCA and optimizing its computation for efficient data analysis.
E N D
IBM - CVUT Student Research Projects Principal Components Analysis Jan Sobotka (sobotj1@fel.cvut.cz)
PCA Algorithm • Statistical technique • PCA was invented in 1901 by Karl Pearson • Is a technique used to reduce multidimensional data sets to lower dimensions • PCA is mathematically defined as an orthogonal linear transformation that transforms the data to a new coordinate system IBM - CVUT Student Research Projects 2
PCA Usage • Useful statistical technique • Data compression • Computer Vision • Face recognition • Finding patterns in data IBM - CVUT Student Research Projects 3
PCA in six steps • Get some data • Substract the mean • Calculate covariance matrix • Calculate the eigenvectors and eigenvalues of the covariance matrix • Choosing components and forming a feature vector • Deriving the new data set IBM - CVUT Student Research Projects 4
My implementation • Obtain information and sources about Cell and PCA • Write code for PPU unit • Port kernel of computation to one SPU unit • Optimize computation on one SPU IBM - CVUT Student Research Projects 5
IBM - CVUT Student Research Projects Thank you for your attention