Coadd: Difference between revisions
imported>Jeremy (Importing text file) |
imported>Jeremy (Importing text file) |
||
| Line 1: | Line 1: | ||
===Purpose=== | ===Purpose=== | ||
Reduce resolution through combination of adjacent variables or samples. | Reduce resolution through combination of adjacent variables or samples. | ||
===Synopsis=== | ===Synopsis=== | ||
:databin = coadd(data,bins,''options'') | :databin = coadd(data,bins,''options'') | ||
:databin = coadd(data,bins,''dim'') | :databin = coadd(data,bins,''dim'') | ||
===Description=== | ===Description=== | ||
COADD is used to combine ("bin") adjacent variables, samples, or slabs of a matrix. Inputs include the original array data, the number of elements to combine together bins {default: 2}, and an optional options structure ''options''. Alternatively, the input ''options'' can be replaced with a scalar value of ''dim'' which will be used for options.dim (see below) and all other options will be the default values. | COADD is used to combine ("bin") adjacent variables, samples, or slabs of a matrix. Inputs include the original array data, the number of elements to combine together bins {default: 2}, and an optional options structure ''options''. Alternatively, the input ''options'' can be replaced with a scalar value of ''dim'' which will be used for options.dim (see below) and all other options will be the default values. | ||
The mode of co-adding (defined by the options value mode) defines how items within each bin are combined mathematically. See options below for details. | The mode of co-adding (defined by the options value mode) defines how items within each bin are combined mathematically. See options below for details. | ||
Unpaired values at the end of the matrix are padded with the least biased value to complete the bin. Output is the co-added data. Unlike DERESOLV, COADD reduces the size of the data matrix by a factor of 1/bins for the dimension. | Unpaired values at the end of the matrix are padded with the least biased value to complete the bin. Output is the co-added data. Unlike DERESOLV, COADD reduces the size of the data matrix by a factor of 1/bins for the dimension. | ||
===Example=== | ===Example=== | ||
Given a matrix, data, size 300 by 1000, the following would coadd variables in groups of three: | Given a matrix, data, size 300 by 1000, the following would coadd variables in groups of three: | ||
:databin = coadd(data,3); | :databin = coadd(data,3); | ||
and the following would coadd samples in groups of two: | and the following would coadd samples in groups of two: | ||
:options.dim = 1; | :options.dim = 1; | ||
:databin = coadd(data,2,options); | :databin = coadd(data,2,options); | ||
The following is equivalent to the previous two lines using the "shortcut" input of dim. | The following is equivalent to the previous two lines using the "shortcut" input of dim. | ||
:databin = coadd(data,2,1); | :databin = coadd(data,2,1); | ||
===Options=== | ===Options=== | ||
* '''dim''': Dimension in which to do combination {default = 2}, | * '''dim''': Dimension in which to do combination {default = 2}, | ||
* '''mode''': [ 'sum' | {'mean'} | 'prod' ] method of combination. See algorithm notes for details of these modes. | * '''mode''': [ 'sum' | {'mean'} | 'prod' ] method of combination. See algorithm notes for details of these modes. | ||
===Algorithm=== | ===Algorithm=== | ||
The three modes, sum, mean and prod behave according to the following (described in terms of variables): | The three modes, sum, mean and prod behave according to the following (described in terms of variables): | ||
SUM: groups of variables are added together and stored. The resulting values will be larger in magnitude than the original values by a factor equal to the number of variables binned. | SUM: groups of variables are added together and stored. The resulting values will be larger in magnitude than the original values by a factor equal to the number of variables binned. | ||
MEAN: groups of variables are added together and that sum is divided by the number of variables binned. The resulting values will be similar in magnitude to the original values. | MEAN: groups of variables are added together and that sum is divided by the number of variables binned. The resulting values will be similar in magnitude to the original values. | ||
PROD: groups of variables are multiplied together. | PROD: groups of variables are multiplied together. | ||
===See Also=== | ===See Also=== | ||
[[deresolv]] | [[deresolv]] | ||
Revision as of 14:24, 3 September 2008
Purpose
Reduce resolution through combination of adjacent variables or samples.
Synopsis
- databin = coadd(data,bins,options)
- databin = coadd(data,bins,dim)
Description
COADD is used to combine ("bin") adjacent variables, samples, or slabs of a matrix. Inputs include the original array data, the number of elements to combine together bins {default: 2}, and an optional options structure options. Alternatively, the input options can be replaced with a scalar value of dim which will be used for options.dim (see below) and all other options will be the default values.
The mode of co-adding (defined by the options value mode) defines how items within each bin are combined mathematically. See options below for details.
Unpaired values at the end of the matrix are padded with the least biased value to complete the bin. Output is the co-added data. Unlike DERESOLV, COADD reduces the size of the data matrix by a factor of 1/bins for the dimension.
Example
Given a matrix, data, size 300 by 1000, the following would coadd variables in groups of three:
- databin = coadd(data,3);
and the following would coadd samples in groups of two:
- options.dim = 1;
- databin = coadd(data,2,options);
The following is equivalent to the previous two lines using the "shortcut" input of dim.
- databin = coadd(data,2,1);
Options
- dim: Dimension in which to do combination {default = 2},
- mode: [ 'sum' | {'mean'} | 'prod' ] method of combination. See algorithm notes for details of these modes.
Algorithm
The three modes, sum, mean and prod behave according to the following (described in terms of variables):
SUM: groups of variables are added together and stored. The resulting values will be larger in magnitude than the original values by a factor equal to the number of variables binned.
MEAN: groups of variables are added together and that sum is divided by the number of variables binned. The resulting values will be similar in magnitude to the original values.
PROD: groups of variables are multiplied together.