RP_maker_diagnose

../_images/smdrqa_logo1.svg

RP_maker_diagnose#

Features#

fnnhitszero_Plot()#

fnnhitszero_Plot(u, n, d, m, tau, sig, delta, Rmin, Rmax, rdiv)[source]#

Function that calculates the ratio of false nearest neighbours

Parameters:

  • u (ndarray) – double array of shape (n,d). Think of it as n points in a d dimensional space

  • n (int) – number of samples or observations in the time series

  • d (int) – number of measurements or dimensions of the data

  • tau (int) – amount of delay

  • m (int) – number of embedding dimensions

  • r (double) – ratio parameter

  • sig (double) – standard deviation of the data

  • delta (double) – the tolerance value(the maximum difference from zero) for a value of FNN ratio to be effectively considered to be zero

  • Rmin (double) – minimum value of r from where we would start the parameter search

  • Rmax (double) – maximum value of r for defining the upper limit of parameter search

  • rdiv (Int) – number of divisions between Rmin and Rmax for parameter search

Returns:

  • Rarr (array) – an array of r values

  • FNN (array) – corresponding false nearest neighbour values

References

  • Kennel, M. B., Brown, R., & Abarbanel, H. D. (1992). Determining embedding dimension for phase-space reconstruction using a geometrical construction. Physical review A, 45 (6), 3403.

findm_Plot()#

findm_Plot(u, n, d, tau, sd, delta, Rmin, Rmax, rdiv, bound, save_path)[source]#

This is an effort to make plot given in Kantz, & Schreiber(2004) section 3.3.1

Parameters:

  • u (ndarray) – double array of shape (n,d). Think of it as n points in a d dimensional space

  • n (int) – number of samples or observations in the time series

  • d (int) – number of measurements or dimensions of the data

  • tau (int) – amount of delay

  • r (double) – ratio parameter

  • sig (double) – standard deviation of the data

  • delta (double) – the tolerance value(the maximum difference from zero) for a value of FNN ratio to be effectively considered to be zero

  • Rmin (double) – minimum value of r from where we would start the parameter search

  • Rmax (double) – maximum value of r for defining the upper limit of parameter search

  • rdiv (Int) – number of divisions between Rmin and Rmax for parameter search

  • bound (double) – bound value for terminating the parameter serch for m

Returns:

Saves a figure and a pickle file

References

  • Kennel, M. B., Brown, R., & Abarbanel, H. D. (1992). Determining embedding dimension for phase-space reconstruction using a geometrical construction. Physical review A, 45 (6), 3403.

  • Kantz, H., & Schreiber, T. (2004). Nonlinear time series analysis (Vol. 7). Cambridge university press. section 3.3.1

RP_diagnose()#

RP_diagnose(input_path, diagnose_dir, rdiv=451, Rmin=1, Rmax=10, delta=0.001, bound=0.2)[source]#

Function to diagnose issues in finding the embedding dimension. It is similar to RP maker, but it deos not generate RP, nstead saves r vs FNN plot varying embedding dimensions and such plots are saved for each of the time series files present in the input directory

Parameters:

  • input_path (str) – folder containing the numpy files, rows> number of samples, columns> number of streams

  • diagnose_dir (str) – folder where the plots needed for checks should be saved

  • rdiv (int) – number of divisions(resolution) for the variable r during parameter search for embedding dimension

  • Rmax (double) – maximum value for the variable r during parameter search for embedding dimension

  • Rmin (double) – minimum value for the variable r during parameter search for embedding dimension

  • delta (double) – the tolerance value below which an FNN value will be considered as zero

  • bound (double) – This is the value in the r value(at which FNN hits zero) va embedding dimension plot. The search is terminated if the value goes below this tolerance value and the value just below tolerance value is reported for embedding dimmension

Returns:

  • Saves r vs FNN plot varying embedding dimensions and such plots are saved for each of the time series files present in the input directory to a path specified as diagnose directory

  • Error_Report_Sheet (file) – This is a csv file containing details of the files for which RP calculation was failed because of numpy.core._exceptions.MemoryError. This is due to an issue at the time delay estimation part, check dimensionality of the data

References

  • Kennel, M. B., Brown, R., & Abarbanel, H. D. (1992). Determining embedding dimension for phase-space reconstruction using a geometrical construction. Physical review A, 45 (6), 3403.

  • Kantz, H., & Schreiber, T. (2004). Nonlinear time series analysis (Vol. 7). Cambridge university press. section 3.3.1

get_minFNN_distribution_plot()#

get_minFNN_distribution_plot(path, save_name)[source]#

This is a function used to get the delta value or the value used to effectively consider a particular value of FNN effectively as zero. This function estimates and plots the distribution of minimum FNN value for different embedding dimensions(m). It computes the upper(2.5% quantile) and lower(97.5% quantile) and when we want to get r(at FNN hitting zero) vs m graph, generally the delta value should be more than the highest upper bound(most likely for m=1) is choosen

Parameters:

  • path (str) – path to folder containing pickes files computed using “RP_diagnose” function

  • savename (str) – The output plot and CSV file would be saved in the name specified

Returns:

Saves a plot and a CSV file

References

  • Kennel, M. B., Brown, R., & Abarbanel, H. D. (1992). Determining embedding dimension for phase-space reconstruction using a geometrical construction. Physical review A, 45 (6), 3403.