This page was created by the IDL library routine
mk_html_help2.
Last modified: Wed Dec 20 10:37:44 2023.
FUNCTION average(array,d [,STDEV=stdev] [,/NAN]) PURPOSE: Returns the average value of an array. The input array can be an array of structures Similar to TOTAL, but returns the average over the given dimension. Also returns standard deviation via an optional keyword argument. Works with structures only if d eq 0 Notes: $LastChangedBy: ali $ $LastChangedDate: 2017-05-01 12:37:30 -0700 (Mon, 01 May 2017) $ $LastChangedRevision: 23254 $ $URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/average.pro $
(See general/tools/misc/average.pro)
FUNCTION average_hist(d,x [,STDEV=stdev])
returns the average of d binned according to x
See "HISTBINS" for a description of keywords
USAGE:
assuming:
x = randomu(seed,1000)*10
y = 10-.1*x^2 + randomn(seed,1000)
d = y
avg = average_hist(d,x,xbins=xc)
avg = average_hist(d,x,xbins=xc,range=[2,8],binsize=.25)
plot,x,y,psym=3
oplot,xc,avg,psym=-4
NOTE: d can be an array of structures:
d=replicate({x:0.,y:0.},1000)
d.x = x
d.y = y
plot,d.x,d.y,psym=3
avg = average_hist(d,d.x)
oplot,avg.x,avg.y,psym=-4
(See general/tools/misc/average_hist.pro)
FUNCTION: dgen(n)
PURPOSE: returns an array of n doubles that are scaled between two limits.
INPUT: n: number of data points. (uses 100 if no value is passed)
KEYWORDS: one of the next 3 keywords should be set:
XRANGE: uses !x.crange (current x limits) for the scaling.
YRANGE: uses !y.crange (current y limits) for the scaling.
RANGE: user selectable range.
LOG: user selectable log scale (Used with RANGE)
EXAMPLES:
x = dgen(/x) ; Returns 100 element array of points evenly distributed along
; the x-axis.
(See general/tools/misc/dgen.pro)
Written by Davin Larson - August 2016
$LastChangedBy: davin-mac $
$LastChangedDate: 2023-11-10 22:27:01 -0800 (Fri, 10 Nov 2023) $
$LastChangedRevision: 32230 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/dynamicarray__define.pro $
Purpose: Object that provides an efficient means of concatenating arrays
da= DynamicArray([InitialArray][,name='name1')
Purpose: Returns a "dynamic array" object. This dynamic array can have any number of elements and can be efficiently
appended to.
This routine is particularly useful when appending to large arrays on numerous occassions. This is especially useful
when the final size of the array is not known when first initialized.
It is functionally equivalent to:
a= findgen(10)
b = 1.
a = [a,b]
but considerably more effiecient because the array size does not need to be increased at every append operation.
Because one can produce arrays of objects, it is a conveniant way of constructing arrays of arrays
containing different things.
Works with multidimensional arrays too.
USAGE:
da = dynamicarray(findgen(1000000), name='Test1')
Or
da.array = findgen(1000000) ; equivalent
Or
da = dynamicarray() & da.append, findgen(1000000)
da.append, findgen(1000) ; append some data
da.append, !values.f_nan ; add a NAN at the end.
da.name = 'NewName' ; change name
a = da.array ; retrieve a copy of the array
print,da.size ; print the number of elements in the array (first dimension of multidimensional arrays)
help,da.name ; display the optional user name
object_destroy, da ; cleanup when done.
(See general/tools/misc/dynamicarray__define.pro)
FUNCTION:
w = enclosed(x,y [cx,cy],NCIRCS=NCIRCS,COUNT=COUNT)
PURPOSE:
Returns the indices of a set of x,y points that are inside a contour.
INPUT: x,y: data set of points. (x and y must be the same dimension)
cx,cy: vector of x,y pairs that describe a closed contour.
if cx,cy are not provided then the cursor is used to obtain it.
OUTPUT:
W: Array of indices of x (& y) that are within the contour cx,cy.
NCIRCS: Same dimension as x (& y); integer array giving the number
of times each point is encircled.
COUNT: Size of array W
(See general/tools/misc/enclosed.pro)
WIDGET Procedure:
EXEC
PURPOSE:
Widget tool that executes a user specified routine. This tool runs in the background.
Author:
Davin Larson - Feb 2012
$LastChangedBy: $
$LastChangedDate: $
$LastChangedRevision: $
$URL: $
(See general/tools/misc/exec.pro)
NAME: getxy PURPOSE: Routine that uses the cursor to select points.
(See general/tools/misc/getxy.pro)
PROCEDURE: print_struct, data, tags=tags PURPOSE: prints data in an array of structures. CALLING PROCEDURE: print_struct, data KEYWORDS: TAGS: tagnames of structure to print CREATED BY: Davin Larson, 1997
(See general/tools/misc/print_struct.pro)
FUNCTION: data=read_asc(filename)
PURPOSE:
Reads data from an ascii file and puts data in an array of structures.
Columns of data should be delimited by spaces.
Data is returned as an array of structures. The elements of the structure
correspond to the columns of the file.
CALLING PROCEDURE:
read_ascii,data,'file.dat'
KEYWORDS:
TAGS: If set then the labels in the text line
preceeding the data will be used for the default struct tag names.
(There should be one label per column of data)
FORMAT: a structure that specifies the output format
of the data. For example if the input file has the
following data:
Year Day secs Vx Vy Vz N
1996 123 13.45 512.3 -10.3 10.5 5.3
the format could be specified as:
FORMAT={year:0,day:0,sec:0.d,v:fltarr(3),n:0.}
if this keyword is not specified then a default structure will be created.
NHEADER: set to number of header lines
CREATED BY: Davin Larson
$LastChangedBy: davin-mac $
$LastChangedDate: 2021-11-08 16:33:53 -0800 (Mon, 08 Nov 2021) $
$LastChangedRevision: 30411 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/read_asc.pro $
(See general/tools/misc/read_asc.pro)
WIDGET Procedure:
RECORDER
PURPOSE:
Widget tool that opens a socket and records streaming data from a server (host) and can save it to a file
or send to a user specified routine. This tool runs in the background.
Keywords:
SET_FILE_TIMERES : defines how often the current output file will be closed and a new one will be opened
DIRECTORY: string prepended to fileformat when opening an output file.
Author:
Davin Larson - April 2011
$LastChangedBy: ali $
$LastChangedDate: 2021-09-24 16:07:57 -0700 (Fri, 24 Sep 2021) $
$LastChangedRevision: 30319 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/recorder.pro $
(See general/tools/misc/recorder.pro)
NAME:
SPLINECOEFF
PURPOSE:
This procedure computes coefficients of cubic splines for a
given observational set and smoothing parameter lambda. The
method is coded after Pollock D.S.G. (1999), "A Handbook of
Time-Series Analysis, Signal Processing and Dynamics, Academic
Press", San Diego.
AUTHOR:
Nikola Vitas
Instituto de Astrofísica de Canarias (IAC)
C/ Vía Láctea, s/n
E38205 - La Laguna (Tenerife), España
Email: n.vitas@iac.es
Homepage: nikolavitas.blogspot.com
CATEGORY:
Numerics, statistics.
CALLING SEQUENCE:
coeffs = SPLINECOEFF([x,] y, [sigma=sigma,] lambda=lambda)
OPTIONAL INPUTS:
OUTPUTS:
INPUT KEYWORDS:
LAMBDA = Scalar, float. Smoothing parameter (It can be determined
empirically, by the LS method or by cross- validation, e.g. see
the book of Pollock.) Lambda equals 0 results in a cubic spline
interpolation. In the other extreme, for a very large lambda,
the result is smoothing by a linear function.
SIGMA = Array. Weight of each data point. If not specified, all
measurements have same weight.
OUTPUT KEYWORDS:
X = Array. Independent variable.
Y = Array. Function to be smoothed or interpolated.
EXAMPLE:
x = .....
y = .....
coeffs = SPLINECOEFF(x, y, lambda = 1.d5)
y1 = FLTARR(N_ELEMENTS(y) - 1)
x1 = x[0:N_ELEMENTS(y)-2]
FOR i = 0, N_ELEMENTS(y)-2 DO y1[i] = coeff.d[I] + $
coeff.c[I] * (x[I+1]-x[I]) + $
coeff.b[I] * (x[I+1]-x[I])^2 + $
coeff.a[I] * (x[I+1]-x[I])^3
PLOT, x, y, psym = 3
OPLOT, x1, y1
DEPENDENCIES:
MODIFICATION HISTORY:
Written by Nikola Vitas, December 2005.
- Small modifications, () replaced with [], NV, October, 2012
(See general/tools/misc/spline_smooth.pro)
FUNCTION: spline_smooth
PURPOSE:
Uses spline smoothing to create a smooth curve through noisy data.
This routine uses splinecoeff.pro, written by Nikola Vitas. See
detailed documentation below (and https://github.com/nikolavitas).
USAGE:
ys = spline_smooth(x, y, weight=w, labmda=lambda)
INPUTS:
x: Independent variable.
y: Dependent variable. Same number of elements as x.
KEYWORDS:
WEIGHT: Weights for y. Default = replicate(1.,n_elements(y))
LAMBDA: Smoothing factor. A value of 0 corresponds to cubic
splines. Larger values increase the smoothing.
$LastChangedBy: dmitchell $
$LastChangedDate: 2023-06-23 12:31:58 -0700 (Fri, 23 Jun 2023) $
$LastChangedRevision: 31907 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/spline_smooth.pro $
CREATED BY: David L. Mitchell 2021-10-18
(See general/tools/misc/spline_smooth.pro)
PROCEDURE: tmean
PURPOSE:
Calculate the mean, median, and standard deviation of a 1-D or 2-D
tplot variable over a specified time range. The variable and time range
are selected with the cursor or via keyword. Skew and kurtosis are also
calculated:
skewness: = 0 -> distribution is symmetric about the maximum
< 0 -> distribution is skewed to the left
> 0 -> distribution is skewed to the right
kurtosis: = 0 -> distribution is peaked like a Gaussian
< 0 -> distribution is less peaked than a Gaussian
> 0 -> distribution is more peaked than a Gaussian
This routine can optionally perform cluster analysis to divide the data
into two groups (Jenks natural breaks optimization). Statistics are given
for each group separately.
USAGE:
tmean, var
INPUTS:
var: Tplot variable name or number. If not specified, determine
based on which panel the mouse is in when clicked.
If the variable has two dimensions in y (time and some other
parameter), then you must specify which indices of the second
dimension to calculate statistics for. Data can be either
summed or averaged over the second dimension. See keywords
IND and AVG.
The variable cannot be compound (list of variables to plot
in the same panel). You must specify which variable in the
list you are interested in.
KEYWORDS:
TRANGE: Use this time range instead of getting it interactively
with the cursor. In this case, you must specify var.
IND: If y has two dimensions (time and some other parameter), this
keyword specifies the indices of the second dimension to
calculate statistics for. No default.
AVG: If IND is set and this keyword is also set, then average over
the second dimension. Otherwise, sum over the second dimension.
NaN's are treated as missing data (see MEAN and TOTAL).
ZERO: Treat NaN's as zeroes. This affects the mean but not the sum.
OFFSET: Value to subtract from the data before calculating statistics.
Default = 0.
OUTLIER: Ignore values more than OUTLIER sigma from the mean.
Default = infinity.
MINPTS: If OUTLIER is set, this specifies the minimum number of
points remaining after discarding outliers. Default = 3.
CLUSTER: Perform 1-D cluster analysis to separate the data into two
groups. Statistics are calculated for each cluster separately.
Diagnostics of the quality of the cluster separation are also
provided (see keyword DIAG). Disables OUTLIER.
MAXDZ: Use largest break between clusters near minimum variance
to divide the clusters. Default = 1.
RESULT: Named variable to hold the result.
HIST: Plot a histogram of the distribution in a separate window.
KEEP: Keep the last histogram window open on exit.
NBINS: If HIST is set, number of bins in the histogram.
NPTS: If this is set, then statistics are calculated for NPTS centered
on the time nearest the cursor when clicked (as opposed to
selecting a time range with two clicks).
DST: Retain the distribution. Does not allow compiling multiple
results.
T0: Times in cluster 0.
T1: Times in cluster 1.
DIAG: Return outlier and cluster analysis diagnostics:
minvar : minimum total variance for both clusters
maxvar : maximum total variance for both clusters
maxsep : separation between the clusters
sepval : value of optimal separation between the clusters
edge : distance between sepval and the edge of the distribution
ngud : number of points in the main distribution
nbad : number of outliers
delta : separation* between the core and outliers in SDEV units
-or-
separation# between the clusters in SDEV units
frac : fraction of "bad" points (nbad/(nbad+ngud))
* separation is the distance between the 2-sigma level of the core
distribution and the closest outlier
# separation is the closest distance between the 2-sigma levels of the
clusters
SILENT: Shhh.
$LastChangedBy: dmitchell $
$LastChangedDate: 2023-08-13 15:42:47 -0700 (Sun, 13 Aug 2023) $
$LastChangedRevision: 31994 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/tmean.pro $
CREATED BY: David L. Mitchell
(See general/tools/misc/tmean.pro)
PROCEDURE: topsmooth
PURPOSE:
Calculate a smooth curve through the upper envelope of a tplot variable.
USAGE:
topsmooth, var
INPUTS:
var: Variable name or number.
KEYWORDS:
NPTS: Half-width of boxcar. Default = 16.
FRAC: Fraction of points to ignore in average or interpolates.
Default = 0.5.
INTERP: Interpolate using data that pass through topfilter.
SIGMA: Spline sigma. Default = 1.
LAMBDA: Spline smooth lambda. Default = 1.
SILENT: Shh.
$LastChangedBy: dmitchell $
$LastChangedDate: 2023-06-23 12:31:58 -0700 (Fri, 23 Jun 2023) $
$LastChangedRevision: 31907 $
$URL: svn+ssh://thmsvn@ambrosia.ssl.berkeley.edu/repos/spdsoft/tags/spedas_6_0/general/tools/misc/topsmooth.pro $
CREATED BY: David L. Mitchell
(See general/tools/misc/topsmooth.pro)
PROCEDURE: xy_edit,x,y,bins PURPOSE: Interactively select data points
(See general/tools/misc/xy_edit.pro)
FUNCTION: yymmdd_to_time
PURPOSE:
Returns time (seconds since 1970) given date in format: YYMMDD HHMM
USAGE:
t = yymmdd_to_time(yymmdd [,hhmm])
(yymmdd can be either a long or a string)
Examples:
t = yymmdd_to_time(990421,1422)
print,t,' ',time_string(t)
Created by: Davin Larson, April 1999
(See general/tools/misc/yymmdd_to_time.pro)