Description
The DWT2D function implements Mallat's algorithm (an iterate
algorithm in the Discrete Wavelet Transform field) on 2-dimensional
matrices and applies wavelet transform on multiple sequences
simultaneously.
Usage
td_dwt2d_mle (
data = NULL,
input.columns = NULL,
index.columns = NULL,
range = NULL,
wavelet = NULL,
wavelet.filter = NULL,
level = NULL,
extension.mode = "sym",
compact.output = TRUE,
partition.columns = NULL,
data.sequence.column = NULL,
wavelet.filter.sequence.column = NULL
)
Arguments
data |
Required Argument.
Specifies the name of the tbl_teradata that contains the
sequences to be transformed.
|
input.columns |
Required Argument.
Specifies the names of the columns in the input tbl_teradata
that contains the data to be transformed. These columns must contain
numeric values between -1e308 and 1e308. The function treats NULL as
0.
Types: character OR vector of Strings (character)
|
index.columns |
Required Argument.
Specifies the columns that contain the indexes of the input
sequences. For a matrix, this takes a vector of two elements,
the first element is the column containing the x coordinates
and second element is the column containing the y coordinates.
Types: character OR vector of Strings (character)
|
range |
Optional Argument.
Specifies the start and end indexes of the input data, all of which
must be integers.
The default values for each sequence are:
starty: minimum y index
startx: minimum x index
endy: maximum y index
endx: maximum x index.
Specify the argument value in the format '(startx,starty),(endx,endy)'.
The function treats any NULL value as 0. The
range can specify a maximum of 1,000,000 cells.
Types: character
|
wavelet |
Optional Argument.
Required if 'wavelet.filter' argument is not specified.
Specifies a wavelet filter name. Refers to the supported wavelet filter names
from td_dwt_mle function's 'wavelet' Argument.
Types: character
|
wavelet.filter |
Optional Argument.
Required if 'wavelet' argument is not specified.
Specifies the name of the tbl_teradata that contains the coefficients
of the wavelet filters.
|
level |
Required Argument.
Specifies the wavelet transform level. The value of the
wavelet transform level must be an integer in the range [1, 1000].
Types: numeric
|
extension.mode |
Optional Argument.
Specifies the method for handling border distortion, an
extension mode from the supported extension modes specified below -
td_dwt_mle "extension.mode" argument.
Default Value: "sym"
Permitted Values: sym, zpd, ppd
Types: character
|
compact.output |
Optional Argument.
Specifies whether to ignore rows in which all coefficient values have an
absolute value less than 1e-12. For a sparse input matrix, ignoring such
rows reduces the output table size.
Default Value: TRUE
Types: logical
|
partition.columns |
Optional Argument.
Specifies the names of the partition.columns, which identify the
sequences. Rows with the same partition.columns values belong to the
same sequence. If you specify multiple partition.columns, then the
function treats the first one as the distribute key of the output and
meta tables. By default, all rows belong to one sequence, and the
function generates a distribute key column named dwt_idrandom_name in
both the coefficient table and the meta table. In both tables,
every cell of dwt_idrandom_name has the value 1.
Types: character OR vector of Strings (character)
|
data.sequence.column |
Optional Argument.
Specifies the vector of column(s) that uniquely identifies each row
of the input argument "data". The argument is used to ensure
deterministic results for functions which produce results that vary
from run to run.
Types: character OR vector of Strings (character)
|
wavelet.filter.sequence.column |
Optional Argument.
Specifies the vector of column(s) that uniquely identifies each row
of the input argument "wavelet.filter". The argument is used to
ensure deterministic results for functions which produce results that
vary from run to run.
Types: character OR vector of Strings (character)
|
Value
Function returns an object of class "td_dwt2d_mle" which is a named list
containing Teradata tbl objects.
Named list members can be referenced directly with the "$" operator
using following names:
coefficient
-
meta.table
output
Examples
# Get the current context/connection
con <- td_get_context()$connection
# Load example data.
# dataset contains climate data in cities in California (CA), Texas (TX), and Washington (WA).
# The cities are represented by two-dimensional coordinates (latitude and longitude).
# The data are temperature (in degrees Fahrenheit), pressure (in Mbars),
# and dew point (in degrees Fahrenheit).
loadExampleData("dwt2d_example", "twod_climate_data")
# Create remote tibble objects.
twod_climate_data <- tbl(con, "twod_climate_data")
# Example 1 - The function outputs a coefficient model table and a meta table.
td_dwt2d_out <- td_dwt2d_mle(data = twod_climate_data,
input.columns = c("temp_f","pressure_mbar","dewpoint_f"),
index.columns = c("latitude","longitude"),
wavelet = "db2",
level = 2,
compact.output = TRUE,
partition.columns = c("state")
)