Fichas

Sin título

asd asd Inglés 15 fichas

Fichas del conjunto (15)

1Goal of PCAGoal is to reduce the dimensionality of the data while retaining as much possible of the variation present in the datasheet
2Objective of PCArigdly rotate the axes of this p-dimenstional space to new positions
3Eigenvaluea number telling you how much variance there in the data in that direction
4Indepentent component analysis(ICA) conceptA given signal is generated by linear mixing of independent components. ICA is a statistical analysis method to estiimate those components and mixing rule
5Whiten dataremove any correlations in the data
6Watershed segmentation basicseach grayscale image can be considered as a topograpic map(depth, map, terrain
)
7what is watershed segmentation sensetive to?noise
8watershed segmentation with markersthe effort to provide over segmentation
9CNN aimreduce number of training samples
10which samples are kept after CNN?the samples that are needed to define the decision boundry
11when is SMOTE used?when one class in the training set dominates the other
12DTW basicsalgorithm to measure similarity between two sequences that vary in time or speed
13when are Eculidean distance used in DTWwhen we have extended the sequences (X' and Y')
14DTW restrictions?The path should not go down or to the left. No element should be skipped in a sequence
15how does the the points devides to cells in kNNA Voronoi diagram