Type:	Package
Title:	Time Series with Matrix Profile
Version:	0.4.15
Maintainer:	Francisco Bischoff <fbischoff@med.up.pt>
Description:	A toolkit implementing the Matrix Profile concept that was created by CS-UCR http://www.cs.ucr.edu/~eamonn/MatrixProfile.html.
License:	Apache License (≥ 2.0)
URL:	https://github.com/matrix-profile-foundation/tsmp
BugReports:	https://github.com/matrix-profile-foundation/tsmp/issues
Depends:	R (≥ 4.0)
Imports:	audio, checkmate, doSNOW, foreach, magrittr, parallel, progress, Rcpp, RcppParallel (≥ 5.0.0), RJSONIO
Suggests:	covr, gdtools, knitr, rmarkdown, raster, spelling, testthat (≥ 2.1.0), vdiffr
LinkingTo:	Rcpp (≥ 1.0.3), RcppParallel (≥ 5.0.0)
VignetteBuilder:	knitr
Encoding:	UTF-8
Language:	en-US
LazyData:	true
NeedsCompilation:	yes
RoxygenNote:	7.2.1
SystemRequirements:	GNU make
Packaged:	2022-08-18 16:31:18 UTC; franz
Author:	Francisco Bischoff [aut, cre], Michael Yeh [res, ccp, ctb], Diego Silva [res, ccp, ctb], Yan Zhu [res, ccp, ctb], Hoang Dau [res, ccp, ctb], Michele Linardi [res, ccp, ctb]
Repository:	CRAN
Date/Publication:	2022-08-20 22:00:02 UTC

Pipe operator

Description

See magrittr::%>% for details.

See magrittr::%T>% for details.

Usage

lhs %>% rhs

lhs %T>% rhs

Runs an appropriate workflow based on the parameters passed in.

Description

The goal of this function is to compute all fundamental algorithms on the provided time series data. See details for more information.

Usage

analyze(
  ts,
  windows = NULL,
  query = NULL,
  sample_pct = 1,
  threshold = 0.98,
  n_jobs = 1L
)

Arguments

Details

For now the following is computed:

1. Matrix Profile - exact or approximate based on sample_pct given that a single windows is provided. By default is the exact algorithm;

2. Top 3 Motifs;

3. Top 3 Discords;

4. Plot Matrix Profile, Motifs and Discords.

When windows is not provided or more than a single window is provided, the Pan-Matrix Profile is computed:

1. Compute the upper bound when a threshold is provided (it is, by default);

2. Compute Pan-Matrix Profile for all windows provided, below the upper bound, or a default range when no windows is provided;

3. Top Motifs;

4. Top Discords;

5. Plot Pan-Matrix Profile, motifs and discords.

Value

The appropriate Matrix Profile or Pan-Matrix Profile profile object and also plots the graphics.

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Main API: compute(), discords(), motifs(), visualize()

Examples

# Matrix Profile
result <- analyze(mp_toy_data$data[, 1], 80)

# Pan Matrix Profile
result <- analyze(mp_toy_data$data[, 1])

Convert a TSMP object into another if possible

Description

The base Classes are MatrixProfile and MultiMatrixProfile, but as other functions are used, classes are pushed behind, since the last output normally is the most significant. If you want, for example, to plot the Matrix Profile from a Fluss object, you may use as.matrixprofile() to cast it back.

Usage

as.matrixprofile(.mp)

as.multimatrixprofile(.mp)

as.pmp(.mp)

as.valmod(.mp)

as.fluss(.mp)

as.chain(.mp)

as.discord(.mp)

as.motif(.mp)

as.multimotif(.mp)

as.arccount(.mp)

as.salient(.mp)

Arguments

Value

Returns the object with the new class, if possible.

Functions

• as.matrixprofile(): Cast an object changed by another function back to MatrixProfile.

• as.multimatrixprofile(): Cast an object changed by another function back to MultiMatrixProfile.

• as.pmp(): Cast an object changed by another function back to PMP.

• as.valmod(): Cast an object changed by another function back to MultiMatrixProfile.

• as.fluss(): Cast an object changed by another function back to Fluss.

• as.chain(): Cast an object changed by another function back to Chain.

• as.discord(): Cast an object changed by another function back to Discord.

• as.motif(): Cast an object changed by another function back to Motif.

• as.multimotif(): Cast an object changed by another function back to MultiMotif.

• as.arccount(): Cast an object changed by another function back to ArcCount.

• as.salient(): Cast an object changed by another function back to Salient.

Examples

w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
mp <- find_motif(mp)
class(mp) # first class will be "Motif"

plot(mp) # plots a motif plot

plot(as.matrixprofile(mp)) # plots a matrix profile plot

Corrects the matrix profile using an annotation vector

Description

This function overwrites the current Matrix Profile using the Annotation Vector. Use with caution.

Usage

av_apply(.mp)

Arguments

Value

Returns the input .mp object corrected by the embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_complexity(), av_hardlimit_artifact(), av_motion_artifact(), av_stop_word(), av_zerocrossing()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- av_complexity(mp)
av <- av_apply(mp)

Computes the annotation vector that favors complexity

Description

Computes the annotation vector that favors complexity

Usage

av_complexity(.mp, data, dilution_factor = 0, apply = FALSE)

Arguments

Value

Returns the input .mp object with an embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_apply(), av_hardlimit_artifact(), av_motion_artifact(), av_stop_word(), av_zerocrossing()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
av <- av_complexity(mp, apply = TRUE)

Computes the annotation vector that suppresses hard-limited artifacts

Description

Computes the annotation vector that suppresses hard-limited artifacts

Usage

av_hardlimit_artifact(.mp, data, apply = FALSE)

Arguments

Value

Returns the input .mp object with an embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_apply(), av_complexity(), av_motion_artifact(), av_stop_word(), av_zerocrossing()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
av <- av_hardlimit_artifact(mp, apply = TRUE)

Computes the annotation vector that suppresses motion artifacts

Description

Computes the annotation vector that suppresses motion artifacts

Usage

av_motion_artifact(.mp, data, apply = FALSE)

Arguments

Value

Returns the input .mp object with an embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_apply(), av_complexity(), av_hardlimit_artifact(), av_stop_word(), av_zerocrossing()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
av <- av_motion_artifact(mp, apply = TRUE)

Computes the annotation vector that suppresses stop-word motifs

Description

Computes the annotation vector that suppresses stop-word motifs

Usage

av_stop_word(
  .mp,
  data,
  stop_word_loc,
  exclusion_zone = NULL,
  threshold = 0.1,
  apply = FALSE
)

Arguments

Details

The function is intended to be generic. However, its parameters (stop_word_loc, exclusion_zone and threshold) are highly dataset dependent.

Value

Returns the input .mp object with an embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_apply(), av_complexity(), av_hardlimit_artifact(), av_motion_artifact(), av_zerocrossing()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
av <- av_stop_word(mp, stop_word_loc = 150, apply = TRUE)

Computes the annotation vector that favors number of zero crossing

Description

Computes the annotation vector that favors number of zero crossing

Usage

av_zerocrossing(.mp, data, apply = FALSE)

Arguments

Value

Returns the input .mp object with an embedded annotation vector.

References

• Dau HA, Keogh E. Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17. New York, New York, USA: ACM Press; 2017. p. 125-34.

See Also

Other Annotation vectors: av_apply(), av_complexity(), av_hardlimit_artifact(), av_motion_artifact(), av_stop_word()

Examples

data <- mp_test_data$train$data[1:1000]
w <- 50
mp <- tsmp(data, window_size = w, verbose = 0)
av <- av_zerocrossing(mp, apply = TRUE)

Computes the Matrix Profile or Pan-Matrix Profile

Description

Main API Function

Usage

compute(
  ts,
  windows = NULL,
  query = NULL,
  sample_pct = 1,
  threshold = 0.98,
  n_jobs = 1L
)

Arguments

Details

Computes the exact or approximate Matrix Profile based on the sample percent specified. Currently, MPX and SCRIMP++ are used for the exact and approximate algorithms respectively. See details for more information about the arguments combinations.

When a single windows is given, the Matrix Profile is computed. If a query is provided, AB join is computed. Otherwise the self-join is computed. When multiple windows or none are given, the Pan-Matrix Profile is computed. If a threshold is set (it is, by default), the upper bound will be computed and the given windows or a default range (when no windows), below the upper bound will be computed.

Value

The profile computed.

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Main API: analyze(), discords(), motifs(), visualize()

Examples

# Matrix Profile
result <- compute(mp_toy_data$data[, 1], 80)

# Pan-Matrix Profile
result <- compute(mp_toy_data$data[, 1])

Search for Discord

Description

Search for Discord

Usage

discords(
  profile,
  exclusion_zone = profile$ez,
  k = 3L,
  neighbor_count = 10L,
  radius = 3
)

Arguments

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Main API: analyze(), compute(), motifs(), visualize()

Calculates the distance profile using MASS algorithms

Description

Mueen's Algorithm for Similarity Search is The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance and Correlation Coefficient.

Usage

dist_profile(
  data,
  query,
  ...,
  window_size = NULL,
  method = "v3",
  index = 1,
  k = NULL,
  weight = NULL,
  paa = 1
)

Arguments

Details

This function has several ways to work:

Case 1: You have a small sized query and the data. In this case you only have to provide the first two parameters data and query. Internally the window_size will be get from the query length.

Case 2: You have one or two data vectors and want to compute the join or self-similarity. In this case you need to use the recursive solution. The parameters are data, query, window_size and index. The first iteration don't need the index unless you are starting somewhere else. The query will be the source of a query_window, starting on index, with length of window_size.

The method defines which MASS will be used. Current supported values are: v2, v3, weighted.

Value

Returns the distance_profile for the given query and the last_product for STOMP algorithm and the parameters for recursive call. See details.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
# minimum example, data and query
nn <- dist_profile(ref_data, ref_data[1:w])
distance_profile <- sqrt(nn$distance_profile)

# data and indexed query
nn <- dist_profile(ref_data, ref_data, window_size = w, index = 10)
distance_profile <- sqrt(nn$distance_profile)

# recursive
nn <- NULL

for (i in seq_len(10)) {
  nn <- dist_profile(ref_data, ref_data, nn, window_size = w, index = i)
}

# weighted
weight <- c(rep(1, w / 3), rep(0.5, w / 3), rep(0.8, w / 3)) # just an example

nn <- dist_profile(ref_data, ref_data,
  window_size = w, index = 1, method = "weighted",
  weight = weight
)
distance_profile <- sqrt(nn$distance_profile)

Fast implementation of moving average and moving standard deviation

Description

This function does not handle NA values

Usage

fast_avg_sd(data, window_size, rcpp = FALSE)

Arguments

Value

Returns a list with avg and sd vectors

Fast implementation of moving average

Description

This function does not handle NA values

Usage

fast_movavg(data, window_size)

Arguments

Value

Returns a vector with the moving average

Examples

data_avg <- fast_movavg(mp_toy_data$data[, 1], mp_toy_data$sub_len)

Fast implementation of moving standard deviation

Description

This function does not handle NA values

Usage

fast_movsd(data, window_size, rcpp = FALSE)

Arguments

Value

Returns a vector with the moving standard deviation

Examples

data_sd <- fast_movsd(mp_toy_data$data[, 1], mp_toy_data$sub_len)

Find Time Series Chains

Description

Time Series Chains is a new primitive for time series data mining.

Usage

find_chains(.mp)

Arguments

Value

Returns the input .mp object with a new name chain. It contains: chains, a list of chains found with more than 2 patterns and best with the best one.

References

• Zhu Y, Imamura M, Nikovski D, Keogh E. Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining. Knowl Inf Syst. 2018 Jun 2;1-27.

Website: https://sites.google.com/site/timeserieschain/

Examples

w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
mp <- find_chains(mp)

Search for Discord

Description

Search for Discord

Usage

find_discord(.mp, ...)

## S3 method for class 'MatrixProfile'
find_discord(
  .mp,
  data,
  n_discords = 1,
  n_neighbors = 3,
  radius = 3,
  exclusion_zone = NULL,
  ...
)

## S3 method for class 'PMP'
find_discord(
  .mp,
  data,
  n_discords = 1,
  n_neighbors = 3,
  radius = 3,
  exclusion_zone = NULL,
  ...
)

Arguments

Value

For class MatrixProfile, returns the input .mp object with a new name discord. It contains: discord_idx, a vector of discords found

For class PMP, returns the input .mp object with a new name discord. It contains: discord_idx, a vector of discords found

Examples

# Single dimension data
w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
mp <- find_discord(mp)
pan <- tsmp(mp_gait_data, window_size = 20:30, mode = "pmp")
mp <- find_discord(pan)

Search for Motifs

Description

Search for Motifs

Usage

find_motif(.mp, ...)

## S3 method for class 'MatrixProfile'
find_motif(
  .mp,
  data,
  n_motifs = 3,
  n_neighbors = 10,
  radius = 3,
  exclusion_zone = NULL,
  ...
)

## S3 method for class 'MultiMatrixProfile'
find_motif(
  .mp,
  data,
  n_motifs = 3,
  mode = c("guided", "unconstrained"),
  n_bit = 4,
  exclusion_zone = NULL,
  n_dim = NULL,
  ...
)

## S3 method for class 'PMP'
find_motif(
  .mp,
  data,
  n_motifs = 3,
  n_neighbors = 10,
  radius = 3,
  exclusion_zone = NULL,
  ...
)

Arguments

Value

For class MatrixProfile, returns the input .mp object with a new name motif. It contains: motif_idx, a list of motif pairs found and motif_neighbor a list with respective motif's neighbors.

For class MultiMatrixProfile, returns the input .mp object with a new name motif. It contains: motif_idx, a vector of motifs found and motif_dim a list the dimensions where the motifs were found

For class PMP, returns the input .mp object with a new name motif. It contains: motif_idx, a list of motif pairs found and motif_neighbor a list with respective motif's neighbors.

Examples

# Single dimension data
w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
mp <- find_motif(mp)

# Multidimension data
w <- mp_toy_data$sub_len
data <- mp_toy_data$data[1:200, ]
mp <- tsmp(data, window_size = w, mode = "mstomp", verbose = 0)
mp <- find_motif(mp)
pan <- tsmp(mp_gait_data, window_size = 20:30, mode = "pmp")
mp <- find_motif(pan)

Time Series Snippets: A New Primitive for Time Series Data Mining

Description

Time Series Snippets tries to solve mainly the common problem of summarization "Show me some representative/typical data". As stated by the original paper, potential uses of snippets are: integrating summarizations of files directly into an operating, production of automatically generated reports, for example, summarize a sleep study and also can be used to support a host of higher-level tasks, including the comparison of massive data collections.

Usage

find_snippet(data, s_size, n_snippets = 2L, window_size = s_size/2L)

Arguments

Details

Motifs vs. snippets: While motifs reward fidelity of conservation, snippets also rewards coverage. Informally, coverage is some measure of how much of the data is explained or represented by a given snippet.

Shapelets vs. snippets: shapelets are defined as subsequences that are maximally representative of a class. Shapelets are supervised, snippets are unsupervised. Shapelets are generally biased to be as short as possible. In contrast, we want snippets to be longer, to intuitively capture the "flavor" of the time series.

Value

Returns the snippet : a list of n_snippets snippets fraction : fraction of each snippet snippetidx : the location of each snippet within time series

References

• Imani S, Madrid F, Ding W, Crouter S, Keogh E. Matrix Profile XIII: Time Series Snippets: A New Primitive for Time Series Data Mining. In: 2018 IEEE International Conference on Data Mining (ICDM). 2018.

• Gharghabi S, Imani S, Bagnall A, Darvishzadeh A, Keogh E. Matrix Profile XII: MPdist: A Novel Time Series Distance Measure to Allow Data Mining in More Challenging Scenarios. In: 2018 IEEE International Conference on Data Mining (ICDM). 2018.

Website: https://sites.google.com/site/snippetfinder/

Examples

snippets <- find_snippet(mp_fluss_data$walkjogrun$data[1:300], 40, n_snippets = 2)

snippets <- find_snippet(mp_fluss_data$walkjogrun$data, 120, n_snippets = 3)
plot(snippets)

Fast Low-cost Online Semantic Segmentation (FLOSS)

Description

Fast Low-cost Online Semantic Segmentation (FLOSS)

Usage

floss(
  .mp,
  new_data,
  data_window,
  threshold = 1,
  exclusion_zone = NULL,
  chunk_size = NULL,
  keep_cac = TRUE
)

Arguments

Value

Returns the input .mp object new names: cac the corrected arc count, cac_finalthe combination of cac after repeated calls of floss(), floss with the location of semantic changes and floss_vals with the normalized arc count value of the semantic change positions.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss_extract(), fluss_cac(), fluss_extract(), fluss_score(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
new_data <- mp_fluss_data$tilt_abp$data[1001:1010]
new_data2 <- mp_fluss_data$tilt_abp$data[1011:1020]
w <- 80
mp <- tsmp(data, window_size = w, verbose = 0)
data_window <- 1000
mp <- floss(mp, new_data, data_window)
mp <- floss(mp, new_data2, data_window)

FLOSS - Corrected Arc Counts

Description

Computes the arc count with edge and 'online' correction (CAC).

Usage

floss_cac(.mp, data_window, exclusion_zone = NULL)

Arguments

Details

Original paper suggest using the classic statistical-process-control heuristic to set a threshold where a semantic change may occur in CAC. This may be useful in real-time implementation as we don't know in advance the number of domain changes to look for. Please check original paper (1).

Value

Returns the input .mp object a new name cac with the corrected arc count and cac_final the combination of cac after repeated calls of floss().

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_extract(), floss(), fluss_cac(), fluss_extract(), fluss_score(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
new_data <- mp_fluss_data$tilt_abp$data[1001:1010]
w <- 10
mp <- tsmp(data, window_size = w, verbose = 0)
data_window <- 1000
mp <- stompi_update(mp, new_data, data_window)
mp <- floss_cac(mp, data_window)

FLOSS - Extract Segments

Description

Extract candidate points of semantic changes.

Usage

floss_extract(.mpac, threshold = 1, exclusion_zone = NULL)

Arguments

Value

Returns the input .mp object a new name floss with the location of semantic changes and floss_vals with the normalized arc count value of the semantic change positions.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss(), fluss_cac(), fluss_extract(), fluss_score(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss_cac(mp)
mp <- fluss_extract(mp, 2)

Fast Low-cost Unipotent Semantic Segmentation (FLUSS)

Description

FLUSS is a Domain Agnostic Online Semantic Segmentation that uses the assumption that when few arc are crossing a given index point, means that there is a high probability of semantic change. This function is a wrap to fluss_cac() and fluss_extract().

Usage

fluss(.mp, num_segments = 1, exclusion_zone = NULL)

Arguments

Value

Returns the input .mp object new names: cac, corrected arc count and fluss with the location of semantic changes.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss_extract(), floss(), fluss_cac(), fluss_extract(), fluss_score()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss(mp, 2)

FLUSS - Corrected Arc Counts

Description

Computes the arc count with edge correction (CAC).

Usage

fluss_cac(.mp, exclusion_zone = NULL)

Arguments

Details

Original paper suggest using the classic statistical-process-control heuristic to set a threshold where a semantic change may occur in CAC. This may be useful in real-time implementation as we don't know in advance the number of domain changes to look for. Please check original paper (1).

Value

Returns the input .mp object a new name cac with the corrected arc count.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss_extract(), floss(), fluss_extract(), fluss_score(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss_cac(mp)

FLUSS - Extract Segments

Description

Extract candidate points of semantic changes.

Usage

fluss_extract(.mpac, num_segments = 1, exclusion_zone = NULL)

Arguments

Value

Returns the input .mp object a new name fluss with the location of semantic changes.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss_extract(), floss(), fluss_cac(), fluss_score(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss_cac(mp)
mp <- fluss_extract(mp, 2)

FLUSS - Prediction score calculation

Description

FLUSS - Prediction score calculation

Usage

fluss_score(gtruth, extracted, data_size)

Arguments

Value

Returns the score of predicted semantic transitions compared with the ground truth. Zero is the best, One is the worst.

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Semantic Segmentations: floss_cac(), floss_extract(), floss(), fluss_cac(), fluss_extract(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
truth <- c(945, 875)
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss_cac(mp)
mp <- fluss_extract(mp, 2)
score <- fluss_score(truth, mp$fluss, length(data))

Get the data included in a TSMP object, if any.

Description

Get the data included in a TSMP object, if any.

Usage

get_data(.mp)

Arguments

Value

Returns the data as matrix. If there is more than one series, returns a list.

Examples

mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)
get_data(mp)

Deprecated functions in package tsmp.

Description

The functions listed below are deprecated and will be defunct in the near future. When possible, alternative functions with similar functionality are also mentioned. Help pages for deprecated functions are available at help("-deprecated").

Usage

mass(
  data_fft,
  query_window,
  data_size,
  window_size,
  data_mean,
  data_sd,
  query_mean,
  query_sd
)

mass

For mass, use dist_profile. Original documentation at mass-deprecated.

Calculates the distance profile using MASS_V2 algorithm

Description

Mueen's Algorithm for Similarity Search is The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance and Correlation Coefficient.

Usage

mass(data_fft, query_window, data_size, window_size, data_mean, data_sd,
 query_mean, query_sd)

Arguments

Value

Returns the distance_profile for the given query and the last_product for STOMP algorithm.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_pre() to precomputation of input values.

tsmp-deprecated

Precomputes several values used on MASS

Description

Precomputes several values used on MASS

Usage

mass_pre(data, query = NULL, window_size)

Arguments

Value

Returns window_size, data_fft, data_size, data_mean, data_sd, query_mean and query_sd.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance.

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_v2(), mass_v3() for using precomputed values.

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 1]

pre <- tsmp:::mass_pre(ref_data, query_data, w)

dp <- list()
for (i in 1:(pre$data_size - w + 1)) {
  dp[[i]] <- tsmp:::mass_v2(
    query_data[i:(i - 1 + w)], pre$window_size, pre$data_fft, pre$data_size,
    pre$data_mean, pre$data_sd, pre$query_mean[i], pre$query_sd[i]
  )
}

Precomputes several values used on MASS

Description

Precomputes several values used on MASS

Usage

mass_pre_w(data, query = NULL, window_size, weight)

Arguments

Value

Returns window_size, data_fft, data_size, data_mean, data_sd, data_pre, weight

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance.

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_weighted() for using precomputed values.

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 1]
weight <- c(
  rep(1, mp_toy_data$sub_len / 3), rep(0.5, mp_toy_data$sub_len / 3),
  rep(1, mp_toy_data$sub_len / 3)
)

pre <- tsmp:::mass_pre_w(ref_data, query_data, w, weight)

dp <- list()
for (i in 1:(pre$data_size - w + 1)) {
  dp[[i]] <- tsmp:::mass_weighted(
    query_data[i:(i - 1 + w)], pre$window_size, pre$data_fft, pre$data_size,
    pre$data_mean, pre$data_sd, pre$data_pre, pre$weight
  )
}

Calculates the distance profile using MASS_V2 algorithm

Description

Mueen's Algorithm for Similarity Search is The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance and Correlation Coefficient.

Usage

mass_v2(
  query_window,
  window_size,
  data_fft,
  data_size,
  data_mean,
  data_sd,
  query_mean,
  query_sd,
  ...
)

Arguments

Value

Returns the distance_profile for the given query and the last_product for STOMP algorithm.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_pre() to precomputation of input values.

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 1]
d_size <- length(ref_data)
q_size <- length(query_data)

pre <- tsmp:::mass_pre(ref_data, query_data, w)

dp <- list()
for (i in 1:(d_size - w + 1)) {
  dp[[i]] <- tsmp:::mass_v2(
    query_data[i:(i - 1 + w)], pre$window_size,
    pre$data_fft, pre$data_size, pre$data_mean, pre$data_sd,
    pre$query_mean[i], pre$query_sd[i]
  )
}

Calculates the distance profile using MASS_V3 algorithm

Description

Mueen's Algorithm for Similarity Search is The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance and Correlation Coefficient.

Usage

mass_v3(
  query_window,
  data,
  window_size,
  data_size,
  data_mean,
  data_sd,
  query_mean,
  query_sd,
  k = NULL,
  ...
)

Arguments

Details

This is a piecewise version of MASS that performs better when the size of the pieces are well aligned with the hardware.

Value

Returns the distance_profile for the given query and the last_product for STOMP algorithm.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_pre() to precomputation of input values.

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 1]
d_size <- length(ref_data)
q_size <- length(query_data)

pre <- tsmp:::mass_pre(ref_data, query_data, w)

dp <- list()
for (i in 1:(d_size - w + 1)) {
  dp[[i]] <- tsmp:::mass_v3(
    query_data[i:(i - 1 + w)], ref_data,
    pre$window_size, pre$data_size, pre$data_mean, pre$data_sd,
    pre$query_mean[i], pre$query_sd[i]
  )
}

Calculates the distance profile using MASS_WEIGHTED algorithm

Description

Mueen's Algorithm for Similarity Search is The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance and Correlation Coefficient.

Usage

mass_weighted(
  query_window,
  window_size,
  data_fft,
  data_size,
  data_mean,
  data_sd,
  data_pre,
  weight,
  ...
)

Arguments

Value

Returns the distance_profile for the given query and the last_product for STOMP algorithm.

References

• Abdullah Mueen, Yan Zhu, Michael Yeh, Kaveh Kamgar, Krishnamurthy Viswanathan, Chetan Kumar Gupta and Eamonn Keogh (2015), The Fastest Similarity Search Algorithm for Time Series Subsequences under Euclidean Distance

Website: https://www.cs.unm.edu/~mueen/FastestSimilaritySearch.html

See Also

mass_pre_w() to precomputation of input values.

Examples

w <- mp_toy_data$sub_len
ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 1]
weight <- c(
  rep(1, mp_toy_data$sub_len / 3), rep(0.5, mp_toy_data$sub_len / 3),
  rep(1, mp_toy_data$sub_len / 3)
)

pre <- tsmp:::mass_pre_w(ref_data, query_data, w, weight)

dp <- list()
for (i in 1:(pre$data_size - w + 1)) {
  dp[[i]] <- tsmp:::mass_weighted(
    query_data[i:(i - 1 + w)], pre$window_size, pre$data_fft, pre$data_size,
    pre$data_mean, pre$data_sd, pre$data_pre, pre$weight
  )
}

Get index of the minimum value from a matrix profile and its nearest neighbor

Description

Get index of the minimum value from a matrix profile and its nearest neighbor

Usage

min_mp_idx(.mp, n_dim = NULL, valid = TRUE)

Arguments

Value

returns a matrix with two columns: the minimum and the nearest neighbor

Examples

w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
min_val <- min_mp_idx(mp)

Search for Motifs

Description

Search for Motifs

Usage

motifs(
  profile,
  exclusion_zone = profile$ez,
  k = 3L,
  neighbor_count = 10L,
  radius = 3
)

Arguments

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Main API: analyze(), compute(), discords(), visualize()

Just a synthetic dataset for testing

Description

Just a synthetic dataset for testing

Usage

motifs_discords_small

Format

A vector with 875 observations

Original data used in the FLUSS paper

Description

Contains two datasets used in FLUSS paper (1), first is TiltABP from (2), and second is WalkJogRun from PAMAP's dataset (3)

Usage

mp_fluss_data

Format

A list containing:

data

one column matrix with the dataset's data

gtruth

a vector with the ground truth of semantic change according to provided dataset

window

window size used in original paper

Source

https://sites.google.com/site/onlinesemanticsegmentation/

http://www.cs.ucr.edu/~eamonn/time_series_data/

References

• Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

• Heldt, T., Oefinger, M.B., Hoshiyama, M. and Mark, R.G., 2003, September. Circulatory response to passive and active changes in posture. In IEEE Computers in Cardiology, 2003 (pp. 263-266).

• Reiss, A. and Stricker, D., 2012. Introducing a new benchmarked dataset for activity monitoring. In 16th International Symposium on Wearable Computers (ISWC), 2012, pages 108-109. IEEE, 2012.

Original data used in the Time Series Chain demo

Description

Original data used in the Time Series Chain demo

Usage

mp_gait_data

Format

A matrix with 904 rows and 1 column with the Y data from an accelerometer

Source

https://sites.google.com/site/timeserieschain/

References

• Zhu Y, Imamura M, Nikovski D, Keogh E. Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining. Knowl Inf Syst. 2018 Jun 2;1-27.

Original data used in the Salient Subsequences demo

Description

This is the Meat dataset from UCR Archive modified for Salient discovery. The original data is mixed with Random Walks and the algorithm must pick only the originals.

Usage

mp_meat_data

Format

original is the original dataset with 60+60 observations mixed with 120 random walks:

data

240 time series with length of 448 each.

labels

label of each time series, -666 means a random walk.

sub_len

size of sliding window.

sub is the original dataset embedded in random walks:

data

One time series with length of 107520.

labels

label of each original data.

labels_idx

starting point where the original data was placed.

sub_len

size of sliding window.

Source

http://www.cs.ucr.edu/~eamonn/time_series_data/

References

• Yeh CCM, Van Herle H, Keogh E. Matrix profile III: The matrix profile allows visualization of salient subsequences in massive time series. Proc - IEEE Int Conf Data Mining, ICDM. 2017;579-88.

• Hu B, Rakthanmanon T, Hao Y, Evans S, Lonardi S, Keogh E. Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL. In: 2011 IEEE 11th International Conference on Data Mining. IEEE; 2011. p. 1086-91.

Website: https://sites.google.com/site/salientsubs/

Original data used in the STDS demo

Description

A synthetic dataset base on TRACE dataset and used as Stress Test to STDS algorithm. The TRACE dataset used here is originally from (1), and the version distributed here is from (2)

Usage

mp_test_data

Format

A list of matrices with 215010 rows and 1 dimension:

train$data

training data

train$label

label for training data

test$data

test data

test$label

label for test data

Source

https://sites.google.com/view/weaklylabeled

http://www.cs.ucr.edu/~eamonn/time_series_data/

References

• Roverso, D., Multivariate temporal classification by windowed wavelet decomposition and recurrent neural networks, in 3rd ANS Int'l Topical Meeting on Nuclear Plant Instrumentation, Control and Human-Machine Interface, vol. 20, Washington, DC, USA, 2000.

• Yeh C-CM, Kavantzas N, Keogh E. Matrix profile IV: Using Weakly Labeled Time Series to Predict Outcomes. Proc VLDB Endow. 2017 Aug 1;10(12):1802-12.

Original data used in the mSTAMP demo

Description

A synthetic dataset with embedded MOTIFs for multidimensional discovery

Usage

mp_toy_data

Format

A list with a matrix with 550 rows and 3 dimensions and an int:

data

data with embedded MOTIFs

sub_len

size of sliding window

Source

https://sites.google.com/view/mstamp/

References

• Yeh CM, Kavantzas N, Keogh E. Matrix Profile VI : Meaningful Multidimensional Motif Discovery.

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

MPdist - Distance between Time Series using Matrix Profile

Description

MPdist is a recently introduced distance measure which considers two time series to be similar if they share many similar subsequences, regardless of the order of matching subsequences. It was demonstrated in that MPdist is robust to spikes, warping, linear trends, dropouts, wandering baseline and missing values, issues that are common outside of benchmark datasets.

Usage

mpdist(
  ref_data,
  query_data,
  window_size,
  type = c("simple", "vector"),
  thr = 0.05
)

Arguments

Details

MPdist returns the distance of two time series or a vector containing the distance between all sliding windows. If argument type is set to vector, the vector is returned.

Value

Returns the distance of two time series or a vector containing the distance between all sliding windows.

References

• Gharghabi S, Imani S, Bagnall A, Darvishzadeh A, Keogh E. Matrix Profile XII: MPdist: A Novel Time Series Distance Measure to Allow Data Mining in More Challenging Scenarios. In: 2018 IEEE International Conference on Data Mining (ICDM). 2018.

Website: https://sites.google.com/site/mpdistinfo/

Examples

ref_data <- mp_toy_data$data[, 1]
qe_data <- mp_toy_data$data[, 2]
qd_data <- mp_toy_data$data[150:200, 1]
w <- mp_toy_data$sub_len

# distance between data of same size
deq <- mpdist(ref_data, qe_data, w)

# distance between data of different sizes
ddiff <- mpdist(ref_data, qd_data, w)

# distance vector between data of different sizes
ddvect <- mpdist(ref_data, qd_data, w, type = "vector")

Fast implementation of MP and MPI for internal purposes, without FFT

Description

Fast implementation of MP and MPI for internal purposes, without FFT

Usage

mpx(
  data,
  window_size,
  query = NULL,
  idx = TRUE,
  dist = c("euclidean", "pearson"),
  n_workers = 1
)

Arguments

Value

Returns MP and MPI

Examples

mp <- mpx(mp_toy_data$data[1:200, 1], window_size = 30)

Multivariate STOMP algorithm Parallel version

Description

Computes the Matrix Profile and Profile Index for Multivariate Time Series.

Usage

mstomp_par(
  data,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  must_dim = NULL,
  exc_dim = NULL,
  n_workers = 2
)

mstomp(
  data,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  must_dim = NULL,
  exc_dim = NULL
)

Arguments

Details

The Matrix Profile, has the potential to revolutionize time series data mining because of its generality, versatility, simplicity and scalability. In particular it has implications for time series motif discovery, time series joins, shapelet discovery (classification), density estimation, semantic segmentation, visualization, rule discovery, clustering etc. The MSTOMP computes the Matrix Profile and Profile Index for Multivariate Time Series that is meaningful for multidimensional MOTIF discovery. It uses the STOMP algorithm that is faster than STAMP but lacks its anytime property.

Although this functions handles Multivariate Time Series, it can also be used to handle Univariate Time Series. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Value

Returns a MultiMatrixProfile object, a list with the matrix profile mp, profile index pi left and right matrix profile lmp, rmp and profile index lpi, rpi, window size w, number of dimensions n_dim, exclusion zone ez, must dimensions must and excluded dimensions exc.

If the input has only one dimension, returns the same as stomp().

Functions

• mstomp_par(): Parallel version.

• mstomp(): Single thread version.

References

• Yeh CM, Kavantzas N, Keogh E. Matrix Profile VI : Meaningful Multidimensional Motif Discovery.

• Zhu Y, Imamura M, Nikovski D, Keogh E. Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining. Knowl Inf Syst. 2018 Jun 2;1-27.

Website: https://sites.google.com/view/mstamp/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: scrimp(), stamp_par(), stomp_par(), tsmp(), valmod()

Examples

# using all dimensions
mp <- mstomp(mp_toy_data$data[1:150, ], 30, verbose = 0)

#' # using threads
mp <- mstomp_par(mp_toy_data$data[1:150, ], 30, verbose = 0)

# force using dimensions 1 and 2
mp <- mstomp(mp_toy_data$data[1:200, ], 30, must_dim = c(1, 2))
# exclude dimensions 2 and 3
mp2 <- mstomp(mp_toy_data$data[1:200, ], 30, exc_dim = c(2, 3))

Plot a TSMP object

Description

Plot a TSMP object

Usage

## S3 method for class 'ArcCount'
plot(
  x,
  data,
  type = c("data", "matrix"),
  exclusion_zone = NULL,
  edge_limit = NULL,
  threshold = stats::quantile(x$cac, 0.1),
  main = "Arcs Discover",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Valmod'
plot(
  x,
  ylab = "distance",
  xlab = "index",
  main = "Valmod Matrix Profile",
  data = FALSE,
  ...
)

## S3 method for class 'MatrixProfile'
plot(
  x,
  ylab = "distance",
  xlab = "index",
  main = "Unidimensional Matrix Profile",
  data = FALSE,
  ...
)

## S3 method for class 'MultiMatrixProfile'
plot(
  x,
  ylab = "distance",
  xlab = "index",
  main = "Multidimensional Matrix Profile",
  ...
)

## S3 method for class 'SimpleMatrixProfile'
plot(
  x,
  ylab = "distance",
  xlab = "index",
  main = "SiMPle Matrix Profile",
  data = FALSE,
  ...
)

## S3 method for class 'Fluss'
plot(
  x,
  data,
  type = c("data", "matrix"),
  main = "Fast Low-cost Unipotent Semantic Segmentation",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Floss'
plot(
  x,
  data,
  type = c("data", "matrix"),
  main = "Fast Low-cost Online Semantic Segmentation",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Chain'
plot(
  x,
  data,
  type = c("data", "matrix"),
  main = "Chain Discover",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Discord'
plot(
  x,
  data,
  type = c("data", "matrix"),
  ncol = 3,
  main = "Discord Discover",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Snippet'
plot(
  x,
  data,
  ncol = 3,
  main = "Snippet Finder",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Motif'
plot(
  x,
  data,
  type = c("data", "matrix"),
  ncol = 3,
  main = "MOTIF Discover",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'MultiMotif'
plot(
  x,
  data,
  type = c("data", "matrix"),
  ncol = 3,
  main = "Multidimensional MOTIF Discover",
  xlab = "index",
  ylab = "",
  ...
)

## S3 method for class 'Salient'
plot(x, data, main = "Salient Subsections", xlab = "index", ylab = "", ...)

## S3 method for class 'PMP'
plot(
  x,
  ylab = "distance",
  xlab = "index",
  main = "Unidimensional Matrix Profile",
  data = FALSE,
  ...
)

Arguments

Value

None

Examples

mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)
plot(mp)

Plot arcs between indexes of a Profile Index

Description

Sometimes may be useful to see where is the nearest neighbor graphically. This is the reasoning behind, for example, FLUSS which uses the arc count to infer a semantic change, and SiMPle which infer that arcs connect similar segments of a music. See details for a deeper explanation how to use this function.

Usage

plot_arcs(
  pairs,
  alpha = NULL,
  quality = 30,
  lwd = 15,
  col = c("blue", "orange"),
  main = "Arc Plot",
  ylab = "",
  xlab = "Profile Index",
  xmin = NULL,
  xmax = NULL,
  ...
)

Arguments

Details

You have two options to use this function. First you can provide just the data, and the function will try its best to retrieve the pairs for plotting. Second, you can skip the first parameters and just provide the pairs, which is a matrix with two columns; the first is the starting index, the second is the end index. Two colors are used to allow you to identify the direction of the arc. If you use the rpi or lpi as input, you will see that these profile indexes have just one direction.

exclusion_zone is used to filter out small arcs that may be useless (e.g. you may be interested in similarities that are far away). edge_limit is used to filter out spurious arcs that are used connect the beginning and the end of the profile (e.g. silent audio). threshold is used to filter indexes that have distant nearest neighbor (e.g. retrieve only the best motifs).

Value

None

Examples

plot_arcs(pairs = matrix(c(5, 10, 1, 10, 20, 5), ncol = 2, byrow = TRUE))

Pan-Matrix Profile

Description

Computes the Pan-Matrix Profile (PMP) for the given time series.

Usage

pmp(
  data,
  window_sizes = seq.int(from = 10, to = length(data)/2, length.out = 20),
  plot = FALSE,
  pmp_obj = NULL,
  n_workers = 1,
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

The work closest in spirit to ours is VALMOD. The idea of VALMOD is to compute the MP for the shortest length of interest, then use the information gleaned from it to guide a search through longer subsequence lengths, exploiting lower bounds to prune off some calculations. This idea works well for the first few of the longer subsequence lengths, but the lower bounds progressively weaken, making the pruning ineffective. Thus, in the five case studies they presented, the mean value of U/L was just 1.24. In contrast, consider that our termite example in Fig. 15 has a U/L ratio of 240, more than two orders of magnitude larger. Thus, VALMOD is perhaps best seen as finding motifs with some tolerance for a slightly (~25%) too short user-specified query length, rather than a true "motif-of-all-lengths" algorithm. Also note that apart from the shortest length, VALMOD only gives some information for the other lengths, unlike pmp, which contains exact distances for all subsequences of all lengths.

When just the data is provided, the exploration will be done using the default window_sizes that is a sequence of 20 values between 10 and the half data size and the resulting object will have an upper_bound equals to Inf. If an object is provided by the argument pmp_obj, this function will add more information to the resulting object, never changing the values already computed. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Talk about upper bound and window sizes

1. upper_window will be set to Inf on new objects 1.1. upper_window will also be used for plot, and for discovery, it must not remove any existing data from the object

2. window_sizes is used for plot, it must not remove any mp inside the object 2.1. window_sizes tells the function what mp are stored, it may be updated with as.numeric(names(pmp))

3. the functions must be capable to handle the data without need to sort by window_size, but sort may be useful later(?)

Value

Returns a PMP object.

Examples

# Just compute
pan <- pmp(mp_gait_data)
# Compute the upper bound, than add new profiles
pan <- pmp_upper_bound(mp_gait_data)
pan <- pmp(mp_gait_data, pmp_obj = pan)

Pan Matrix Profile upper bound

Description

Finds the upper bound for Pan Matrix Profile calculation.

Usage

pmp_upper_bound(
  data,
  threshold = getOption("tsmp.pmp_ub", 0.95),
  refine_stepsize = getOption("tsmp.pmp_refine", 0.25),
  return_pmp = TRUE,
  n_workers = 1,
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

The Pan Matrix Profile may not give any further information beyond a certain window size. This function starts computing the matrix profile for the window size of 8 and doubles it until the minimum correlation value found is less than the threshold. After that, it begins to refine the upper bound using the refine_stepsize values, until the threshold value is hit.

verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Value

Returns a PMP object with computed data, or just the upper bound value if return_pmp is set to FALSE.

References

• Yet to be announced

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

Examples

# return the object
pan_matrix <- pmp_upper_bound(mp_gait_data)

# just the upper bound
pan_ub <- pmp_upper_bound(mp_gait_data, return_pmp = FALSE)

Read TSMP object from JSON file.

Description

Read TSMP object from JSON file.

Usage

read(x, ...)

Arguments

Examples

result <- compute(mp_toy_data$data[, 1], 80)

tempfile <- file.path(tempdir(), "output.json")
write(result, file = tempfile)
result <- read(tempfile)

Remove a TSMP class from an object

Description

Remove a TSMP class from an object

Usage

remove_class(x, class)

Arguments

Value

the object without the class

Examples

w <- 50
data <- mp_gait_data
mp <- tsmp(data, window_size = w, exclusion_zone = 1 / 4, verbose = 0)
mp <- find_chains(mp)
# Remove the "Chain" class information
mp <- remove_class(mp, "Chain")

Convert salient sequences into MDS space

Description

Convert salient sequences into MDS space

Usage

salient_mds(.mp, data, bit_idx = 1)

Arguments

Value

Returns X,Y values for plotting

References

• Yeh CCM, Van Herle H, Keogh E. Matrix profile III: The matrix profile allows visualization of salient subsequences in massive time series. Proc - IEEE Int Conf Data Mining, ICDM. 2017;579-88.

• Hu B, Rakthanmanon T, Hao Y, Evans S, Lonardi S, Keogh E. Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL. In: 2011 IEEE 11th International Conference on Data Mining. IEEE; 2011. p. 1086-91.

Website: https://sites.google.com/site/salientsubs/

Examples

# toy example
data <- mp_toy_data$data[, 1]
mp <- tsmp(data, window_size = 30, verbose = 0)
mps <- salient_subsequences(mp, verbose = 0)
mds_data <- salient_mds(mps)
plot(mds_data, main = "Multi dimensional scale")

Computes the F-Score of salient algorithm.

Description

This score function is useful for testing several values of n_bits for MDL discretization and checking against a known set of indexes. This increase the probability of better results on relevant subsequence extraction.

Usage

salient_score(.mp, gtruth, verbose = getOption("tsmp.verbose", 2))

Arguments

Value

Returns a list with f_score, precision, recall and bits used in the algorithm.

References

• Yeh CCM, Van Herle H, Keogh E. Matrix profile III: The matrix profile allows visualization of salient subsequences in massive time series. Proc - IEEE Int Conf Data Mining, ICDM. 2017;579-88.

• Hu B, Rakthanmanon T, Hao Y, Evans S, Lonardi S, Keogh E. Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL. In: 2011 IEEE 11th International Conference on Data Mining. IEEE; 2011. p. 1086-91.

Website: https://sites.google.com/site/salientsubs/

Examples

# toy example
data <- mp_toy_data$data[, 1]
mp <- tsmp(data, window_size = 30, verbose = 0)
mps <- salient_subsequences(mp, n_bits = c(4, 6, 8), verbose = 0)
label_idx <- seq(2, 500, by = 110) # fake data
salient_score(mps, label_idx, verbose = 0)

Framework for retrieve salient subsequences from a dataset

Description

In order to allow a meaningful visualization in Multi-Dimensional Space (MDS), this function retrieves the most relevant subsequences using Minimal Description Length (MDL) framework.

Usage

salient_subsequences(
  .mp,
  data,
  n_bits = 8,
  n_cand = 10,
  exclusion_zone = NULL,
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Value

Returns the input .mp object with a new name salient. It contains: indexes, a vector with the starting position of each subsequence, idx_bit_size, a vector with the associated bitsize for each iteration and bits the value used as input on n_bits.

References

• Yeh CCM, Van Herle H, Keogh E. Matrix profile III: The matrix profile allows visualization of salient subsequences in massive time series. Proc - IEEE Int Conf Data Mining, ICDM. 2017;579-88.

• Hu B, Rakthanmanon T, Hao Y, Evans S, Lonardi S, Keogh E. Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL. In: 2011 IEEE 11th International Conference on Data Mining. IEEE; 2011. p. 1086-91.

Website: https://sites.google.com/site/salientsubs/

Examples

# toy example
data <- mp_toy_data$data[, 1]
mp <- tsmp(data, window_size = 30, verbose = 0)
mps <- salient_subsequences(mp, data, verbose = 0)

# full example
data <- mp_meat_data$sub$data
w <- mp_meat_data$sub$sub_len
mp <- tsmp(data, window_size = w, verbose = 2, n_workers = 2)
mps <- salient_subsequences(mp, data, n_bits = c(4, 6, 8), verbose = 2)

Anytime univariate SCRIMP++ algorithm

Description

Computes the best so far Matrix Profile and Profile Index for Univariate Time Series. DISCLAIMER: This algorithm still in development by its authors. Join similarity, RMP and LMP not implemented yet.

Usage

scrimp(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  s_size = Inf,
  pre_scrimp = 1/4,
  pre_only = FALSE
)

Arguments

Details

The Matrix Profile, has the potential to revolutionize time series data mining because of its generality, versatility, simplicity and scalability. In particular it has implications for time series motif discovery, time series joins, shapelet discovery (classification), density estimation, semantic segmentation, visualization, rule discovery, clustering etc. The anytime SCRIMP computes the Matrix Profile and Profile Index in such manner that it can be stopped before its complete calculation and return the best so far results allowing ultra-fast approximate solutions. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound. exclusion_zone is used to avoid trivial matches.

Value

Returns a MatrixProfile object, a list with the matrix profile mp, profile index pi left and right matrix profile lmp, rmp and profile index lpi, rpi, window size w and exclusion zone ez.

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: mstomp_par(), stamp_par(), stomp_par(), tsmp(), valmod()

Examples

mp <- scrimp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)

ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 2]
# self similarity
mp <- scrimp(ref_data, window_size = 30, s_size = round(nrow(ref_data) * 0.1))
# join similarity
mp <- scrimp(ref_data, query_data, window_size = 30, s_size = round(nrow(query_data) * 0.1))

Framework for Scalable Dictionary learning for Time Series (SDTS) prediction function

Description

This function trains a model that uses a dictionary to predict state changes. Differently from fluss(), it doesn't look for semantic changes (that may be several), but for binary states like "on" or "off". Think for example that a human annotator is pressing a switch any time he thinks that the recorded data is relevant, and releases the switch when he thinks the data is noise. This algorithm will learn the switching points (even better) and try to predict using new data.

Usage

sdts_predict(model, data, window_size)

Arguments

Value

Returns a vector of logical with predicted annotations.

References

• Yeh C-CM, Kavantzas N, Keogh E. Matrix profile IV: Using Weakly Labeled Time Series to Predict Outcomes. Proc VLDB Endow. 2017 Aug 1;10(12):1802-12.

Website: https://sites.google.com/view/weaklylabeled

See Also

Other Scalable Dictionaries: sdts_score(), sdts_train()

Examples

# This is a fast toy example and results are useless. For a complete result, run the code inside
#' Not run' section below.
w <- c(110, 220)
subs <- 11000:20000
tr_data <- mp_test_data$train$data[subs]
tr_label <- mp_test_data$train$label[subs]
te_data <- mp_test_data$test$data[subs]
te_label <- mp_test_data$test$label[subs]
model <- sdts_train(tr_data, tr_label, w, verbose = 0)
predict <- sdts_predict(model, te_data, round(mean(w)))
sdts_score(predict, te_label, 1)

windows <- c(110, 220, 330)
model <- sdts_train(mp_test_data$train$data, mp_test_data$train$label, windows, verbose = 0)
predict <- sdts_predict(model, mp_test_data$test$data, round(mean(windows)))
sdts_score(predict, mp_test_data$test$label, 1)

Computes the F-Score of a SDTS prediction

Description

Computes the F-Score of a SDTS prediction.

Usage

sdts_score(pred, gtruth, beta = 1)

Arguments

Details

beta is used to balance F-score towards recall (⁠>1⁠) or precision (⁠<1⁠).

Value

Returns a list with f_score, precision and recall.

References

• Yeh C-CM, Kavantzas N, Keogh E. Matrix profile IV: Using Weakly Labeled Time Series to Predict Outcomes. Proc VLDB Endow. 2017 Aug 1;10(12):1802-12.

Website: https://sites.google.com/view/weaklylabeled

See Also

Other Scalable Dictionaries: sdts_predict(), sdts_train()

Examples

# This is a fast toy example and results are useless. For a complete result, run the code inside
#' Not run' section below.
w <- c(110, 220)
subs <- 11000:20000
tr_data <- mp_test_data$train$data[subs]
tr_label <- mp_test_data$train$label[subs]
te_data <- mp_test_data$test$data[subs]
te_label <- mp_test_data$test$label[subs]
model <- sdts_train(tr_data, tr_label, w, verbose = 0)
predict <- sdts_predict(model, te_data, round(mean(w)))
sdts_score(predict, te_label, 1)

windows <- c(110, 220, 330)
model <- sdts_train(mp_test_data$train$data, mp_test_data$train$label, windows)
predict <- sdts_predict(model, mp_test_data$test$data, round(mean(windows)))
sdts_score(predict, mp_test_data$test$label, 1)

Framework for Scalable Dictionary learning for Time Series (SDTS) training function

Description

This function trains a model that uses a dictionary to predict state changes. Differently from fluss(), it doesn't look for semantic changes (that may be several), but for binary states like "on" or "off". Think for example that a human annotator is pressing a switch any time he thinks that the recorded data is relevant, and releases the switch when he thinks the data is noise. This algorithm will learn the switching points (even better) and try to predict using new data.

Usage

sdts_train(
  data,
  label,
  window_size,
  beta = 1,
  pat_max = Inf,
  parallel = FALSE,
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

beta is used to balance F-score towards recall (⁠>1⁠) or precision (⁠<1⁠). verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Value

Returns a list with the learned dictionary score (estimated score), score_hist (history of scores), pattern (shape features), thold (threshold values).

References

• Yeh C-CM, Kavantzas N, Keogh E. Matrix profile IV: Using Weakly Labeled Time Series to Predict Outcomes. Proc VLDB Endow. 2017 Aug 1;10(12):1802-12.

Website: https://sites.google.com/view/weaklylabeled

See Also

Other Scalable Dictionaries: sdts_predict(), sdts_score()

Examples

# This is a fast toy example and results are useless. For a complete result, run the code inside
#' Not run' section below.
w <- c(110, 220)
subs <- 11000:20000
tr_data <- mp_test_data$train$data[subs]
tr_label <- mp_test_data$train$label[subs]
te_data <- mp_test_data$test$data[subs]
te_label <- mp_test_data$test$label[subs]
model <- sdts_train(tr_data, tr_label, w, verbose = 0)
predict <- sdts_predict(model, te_data, round(mean(w)))
sdts_score(predict, te_label, 1)

windows <- c(110, 220, 330)
model <- sdts_train(mp_test_data$train$data, mp_test_data$train$label, windows)
predict <- sdts_predict(model, mp_test_data$test$data, round(mean(windows)))
sdts_score(predict, mp_test_data$test$label, 1)

Set/changes the data included in TSMP object.

Description

This may be useful if you want to include the data lately or remove the included data (set as NULL).

Usage

set_data(.mp, data)

Arguments

Value

Returns silently the original TSMP object with changed data.

Examples

mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)
mp <- set_data(mp, NULL)

Compute the join similarity for Sound data

Description

Compute the join similarity for Sound data

Usage

simple_fast(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

Value

Returns a SimpleMatrixProfile object, a list with the matrix profile mp, profile index pi, number of dimensions n_dim, window size w and exclusion zone ez.

References

• Silva D, Yeh C, Batista G, Keogh E. Simple: Assessing Music Similarity Using Subsequences Joins. Proc 17th ISMIR Conf. 2016;23-30.

• Silva DF, Yeh C-CM, Zhu Y, Batista G, Keogh E. Fast Similarity Matrix Profile for Music Analysis and Exploration. IEEE Trans Multimed. 2018;14(8):1-1.

Website: https://sites.google.com/view/simple-fast

Website: https://sites.google.com/site/ismir2016simple/home

Examples

w <- 30
data <- mp_toy_data$data # 3 dimensions matrix
result <- simple_fast(data, window_size = w, verbose = 0)

Anytime univariate STAMP algorithm Parallel version

Description

Computes the best so far Matrix Profile and Profile Index for Univariate Time Series.

Usage

stamp_par(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  s_size = Inf,
  n_workers = 2,
  weight = NULL
)

stamp(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  s_size = Inf,
  weight = NULL
)

Arguments

Details

The Matrix Profile, has the potential to revolutionize time series data mining because of its generality, versatility, simplicity and scalability. In particular it has implications for time series motif discovery, time series joins, shapelet discovery (classification), density estimation, semantic segmentation, visualization, rule discovery, clustering etc. The anytime STAMP computes the Matrix Profile and Profile Index in such manner that it can be stopped before its complete calculation and return the best so far results allowing ultra-fast approximate solutions. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound. exclusion_zone is used to avoid trivial matches; if a query data is provided (join similarity), this parameter is ignored.

Value

Returns a MatrixProfile object, a list with the matrix profile mp, profile index pi left and right matrix profile lmp, rmp and profile index lpi, rpi, window size w and exclusion zone ez.

Functions

• stamp_par(): Parallel version.

• stamp(): Single thread version.

References

• Yeh CCM, Zhu Y, Ulanova L, Begum N, Ding Y, Dau HA, et al. Matrix profile I: All pairs similarity joins for time series: A unifying view that includes motifs, discords and shapelets. Proc - IEEE Int Conf Data Mining, ICDM. 2017;1317-22.

• Zhu Y, Imamura M, Nikovski D, Keogh E. Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining. Knowl Inf Syst. 2018 Jun 2;1-27.

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: mstomp_par(), scrimp(), stomp_par(), tsmp(), valmod()

Examples

mp <- stamp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)


#' # using threads
mp <- stamp_par(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)

ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 2]
# self similarity
mp <- stamp(ref_data, window_size = 30, s_size = round(nrow(ref_data) * 0.1))
# join similarity
mp <- stamp(ref_data, query_data, window_size = 30, s_size = round(nrow(query_data) * 0.1))

Univariate STOMP algorithm

Description

Computes the Matrix Profile and Profile Index for Univariate Time Series.

Usage

stomp_par(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2),
  n_workers = 2
)

stomp(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

The Matrix Profile, has the potential to revolutionize time series data mining because of its generality, versatility, simplicity and scalability. In particular it has implications for time series motif discovery, time series joins, shapelet discovery (classification), density estimation, semantic segmentation, visualization, rule discovery, clustering etc. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound. exclusion_zone is used to avoid trivial matches; if a query data is provided (join similarity), this parameter is ignored.

Value

Returns a MatrixProfile object, a list with the matrix profile mp, profile index pi left and right matrix profile lmp, rmp and profile index lpi, rpi, window size w and exclusion zone ez.

Functions

• stomp_par(): Parallel version.

• stomp(): Single thread version.

References

• Zhu Y, Zimmerman Z, Senobari NS, Yeh CM, Funning G. Matrix Profile II : Exploiting a Novel Algorithm and GPUs to Break the One Hundred Million Barrier for Time Series Motifs and Joins. Icdm. 2016 Jan 22;54(1):739-48.

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: mstomp_par(), scrimp(), stamp_par(), tsmp(), valmod()

Examples

mp <- stomp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)

#' # using threads
mp <- stomp_par(mp_toy_data$data[1:400, 1], window_size = 30, verbose = 0)

ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 2]
# self similarity
mp <- stomp(ref_data, window_size = 30)
# join similarity
mp2 <- stomp(ref_data, query_data, window_size = 30)

Real-time STOMP algorithm

Description

Real-time STOMP algorithm

Usage

stompi_update(.mp, new_data, history_size = FALSE)

Arguments

Value

Returns the input .mp updated with the new information.

Examples

mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)
mpi <- stompi_update(mp, mp_toy_data$data[201:300, 1])
mp <- tsmp(mp_toy_data$data[1:300, 1], window_size = 30, verbose = 0)
all.equal(mp, mpi, check.attributes = FALSE)

Computes the Matrix Profile and Profile Index

Description

This is a wrap function that makes easy to use all available algorithms to compute the Matrix Profile and Profile Index for multiple purposes.

Usage

tsmp(
  ...,
  window_size,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  mode = c("stomp", "stamp", "simple", "mstomp", "scrimp", "valmod", "pmp"),
  verbose = getOption("tsmp.verbose", 2),
  n_workers = 1,
  s_size = Inf,
  must_dim = NULL,
  exc_dim = NULL,
  heap_size = 50,
  paa = 1,
  .keep_data = TRUE
)

Arguments

Details

The Matrix Profile, has the potential to revolutionize time series data mining because of its generality, versatility, simplicity and scalability. In particular it has implications for time series motif discovery, time series joins, shapelet discovery (classification), density estimation, semantic segmentation, visualization, rule discovery, clustering etc.

The first algorithm invented was the stamp() that using mass() as an ultra-fast Algorithm for Similarity Search allowed to compute the Matrix Profile in reasonable time. One of its main feature was its Anytime property which using a randomized approach could return a "best-so-far" matrix that could give us the correct answer (using for example 1/10 of all iterations) almost every time.

The next algorithm was stomp() that currently is the most used. Researchers noticed that the dot products do not need to be recalculated from scratch for each subsequence. Instead, we can reuse the values calculated for the first subsequence to make a faster calculation in the next iterations. The idea is to make use of the intersections between the required products in consecutive iterations. This approach reduced the time to compute the Matrix Profile to about 3% compared to stamp(), but on the other hand, we lost the Anytime property.

Currently there is a new algorithm that I'll not explain further here. It is called scrimp(), and is as fast as stomp(), and have the Anytime property. This algorithm is implemented in this package, but still waiting for an article publication.

Further, there is the mstomp() that computes a multidimensional Matrix Profile that allows to meaningful MOTIF discovery in Multivariate Time Series. And simple_fast() that also handles Multivariate Time Series, but focused in Music Analysis and Exploration.

The valmod() uses a new pruning algorithm allowing a similarity search with a range of sliding window sizes.

The pmp() is a new concept that creates several profiles from a range of windows.

Some parameters are global across the algorithms:

...

One or two time series (except for mstomp()). The second time series can be smaller than the first.

window_size

The sliding window.

exclusion_zone

Is used to avoid trivial matches; if a query data is provided (join similarity), this parameter is ignored.

verbose

Changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound.

n_workers

number of threads for parallel computing (except simple_fast, scrimp and valmod). If the value is 2 or more, the '_par' version of the algorithm will be used.

s_size is used only in Anytime algorithms: stamp() and scrimp(). must_dim and exc_dim are used only in mstomp(). heap_size is used only for valmod() mode can be any of the following: stomp, stamp, simple, mstomp, scrimp, valmod, pmp.

Value

Returns the matrix profile mp and profile index pi. It also returns the left and right matrix profile lmp, rmp and profile index lpi, rpi that may be used to detect Time Series Chains. mstomp() returns a multidimensional Matrix Profile.

References

• Silva D, Yeh C, Batista G, Keogh E. Simple: Assessing Music Similarity Using Subsequences Joins. Proc 17th ISMIR Conf. 2016;23-30.

• Silva DF, Yeh C-CM, Zhu Y, Batista G, Keogh E. Fast Similarity Matrix Profile for Music Analysis and Exploration. IEEE Trans Multimed. 2018;14(8):1-1.

• Yeh CM, Kavantzas N, Keogh E. Matrix Profile VI : Meaningful Multidimensional Motif Discovery.

• Yeh CCM, Zhu Y, Ulanova L, Begum N, Ding Y, Dau HA, et al. Matrix profile I: All pairs similarity joins for time series: A unifying view that includes motifs, discords and shapelets. Proc - IEEE Int Conf Data Mining, ICDM. 2017;1317-22.

• Zhu Y, Imamura M, Nikovski D, Keogh E. Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining. Knowl Inf Syst. 2018 Jun 2;1-27.

• Zhu Y, Zimmerman Z, Senobari NS, Yeh CM, Funning G. Matrix Profile II : Exploiting a Novel Algorithm and GPUs to Break the One Hundred Million Barrier for Time Series Motifs and Joins. Icdm. 2016 Jan 22;54(1):739-48.

Website: https://sites.google.com/view/simple-fast

Website: https://sites.google.com/site/ismir2016simple/home

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: mstomp_par(), scrimp(), stamp_par(), stomp_par(), valmod()

Examples

# default with [stomp()]
mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, verbose = 0)

# Anytime STAMP
mp <- tsmp(mp_toy_data$data[1:200, 1], window_size = 30, mode = "stamp", s_size = 50, verbose = 0)

# [mstomp()]
mp <- tsmp(mp_toy_data$data[1:200, ], window_size = 30, mode = "mstomp", verbose = 0)

# [simple_fast()]
mp <- tsmp(mp_toy_data$data[1:200, ], window_size = 30, mode = "simple", verbose = 0)

# parallel with [stomp_par()]
mp <- tsmp(mp_test_data$train$data[1:1000, 1], window_size = 30, n_workers = 2, verbose = 0)

Variable Length Motif Discovery

Description

Computes the Matrix Profile and Profile Index for a range of query window sizes

Usage

valmod(
  ...,
  window_min,
  window_max,
  heap_size = 50,
  exclusion_zone = getOption("tsmp.exclusion_zone", 1/2),
  lb = TRUE,
  verbose = getOption("tsmp.verbose", 2)
)

Arguments

Details

This algorithm uses an exact algorithm based on a novel lower bounding technique, which is specifically designed for the motif discovery problem. verbose changes how much information is printed by this function; 0 means nothing, 1 means text, 2 adds the progress bar, 3 adds the finish sound. exclusion_zone is used to avoid trivial matches; if a query data is provided (join similarity), this parameter is ignored.

Paper that implements skimp() suggests that window_max / window_min > than 1.24 begins to weakening pruning in valmod().

Value

Returns a Valmod object, a list with the matrix profile mp, profile index pi left and right matrix profile lmp, rmp and profile index lpi, rpi, best window size w for each index and exclusion zone ez. Additionally: evolution_motif the best motif distance per window size, and non-length normalized versions of mp, pi and w: mpnn, pinn and wnn.

References

• Linardi M, Zhu Y, Palpanas T, Keogh E. VALMOD: A Suite for Easy and Exact Detection of Variable Length Motifs in Data Series. In: Proceedings of the 2018 International Conference on Management of Data - SIGMOD '18. New York, New York, USA: ACM Press; 2018. p. 1757-60.

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other matrix profile computations: mstomp_par(), scrimp(), stamp_par(), stomp_par(), tsmp()

Examples

mp <- valmod(mp_toy_data$data[1:200, 1], window_min = 30, window_max = 40, verbose = 0)

ref_data <- mp_toy_data$data[, 1]
query_data <- mp_toy_data$data[, 2]
# self similarity
mp <- valmod(ref_data, window_min = 30, window_max = 40)
# join similarity
mp <- valmod(ref_data, query_data, window_min = 30, window_max = 40)

Plots an object generated from one of the algorithms. In some cases multiple plots will be generated

Description

Plots an object generated from one of the algorithms. In some cases multiple plots will be generated

Usage

visualize(profile)

Arguments

References

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See Also

Other Main API: analyze(), compute(), discords(), motifs()

Examples

result <- compute(mp_toy_data$data[, 1], 80)
visualize(result)

Write a TSMP object to JSON file.

Description

Write a TSMP object to JSON file.

Usage

write(x, ...)

## S3 method for class 'MatrixProfile'
write(x, file, ...)

## S3 method for class 'PMP'
write(x, file, ...)

Arguments

Examples

result <- compute(mp_toy_data$data[, 1], 80)

write(result, file = file.path(tempdir(), "output.json"))