CBF: Cylinder-Bell-Funnel (CBF) data set

In moviedo5/fda.tsc: Functional Data Sets for Time Series Classification

Description

Cylinder-Bell-Funnel is a simulated data set defined by N. Saiko in his thesis "Local Feature Extraction and Its Applications Using a Library of Bases ". Data from each class is starndard normal noise plus an offset term which differs for each class.

Usage

data(CBF)

Format

The variables are as follows:

• df: data.frame with the following variables:

  • class: Corresponding class level of “CBF” curves with 3 classes, 310 observations per class.

  • sample:Factor variable. In TSC database, the first 30 values (sample=train) are used for training sample and the rest of 900 (sample=test) for testing.

• x: fdata class object with with n=930 curves (per row) in 930 discretization points (per column).

Source

http://timeseriesclassification.com/description.php?Dataset=CBF

References

Saito, N., Coifman, R. R. (1994). Local feature extraction and its applications using a library of bases (Doctoral dissertation, Yale University). https://www.math.ucdavis.edu/~saito/publications/saito_phd.pdf

moviedo5/fda.tsc documentation built on May 16, 2019, 12:10 a.m.