Hi Shishir. The algorithms are identical. However the direction of the eigenvectors that generate the new coordinates are only defined up to the sign. So the sign of the axis in both PCA and KPCA is detemined by the details of the algorithm computing the eigenvector. I is not really surprising if they end up in different directions.
Hth, Andy On 02/23/2014 01:25 PM, Shishir Pandey wrote: > Hi > > Just wanted to know what is the difference between PCA and KPCA with > linear kernel. They produce the same result on columns 1, 2, etc. But, > the values in column 0 of the fitted transform of one is negative of > other. Shouldn't they both be the same? > ------------------------------------------------------------------------------ Managing the Performance of Cloud-Based Applications Take advantage of what the Cloud has to offer - Avoid Common Pitfalls. Read the Whitepaper. http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
