Data Visualization

Overview

Last Update: 1 Dec 2006 - updated for CIAO 3.4: ChIPS version

Synopsis:

This thread provides a detailed introduction to data visualization in Sherpa.

Related Links:

Sherpa Configuration: Using the State Objects thread: how the state objects (a.k.a. configuration variables) can be used to customize various features of Sherpa, including plots.
The other threads on plotting in Sherpa.

Proceed to the HTML or hardcopy (PDF: A4 | letter) version of the thread.

Getting Started

Please follow the "Sherpa Threads: Getting Started" thread.

For 1-D data, the primary plotting command is LPLOT. In addition, a FITS Embedded Function (FEF) file that is read in via FEFFILE may be plotted using the command FEFPLOT. And, the OPLOT command may be used to plot multiple data curves into the same drawing area.

This thread demonstrates 1-D data visualization using LPLOT in detail.

Also, see the Sherpa Reference Manual for further information regarding data display capabilities within Sherpa. In particular, the Sherpa Display chapter of the Reference Manual contains information about 1-D data visualization.

1-D Plotting

A Sherpa session typically begins with the input of data. Here, we input a 1-D dataset and errors:

sherpa> READ DATA data1.dat 1 2
sherpa> READ ERRORS data1.dat 1 3

These data may then be plotted from within Sherpa:

sherpa> LPLOT DATA

Figure 1 shows the resulting plot.

Next, we continue the Sherpa session by defining a source model, thawing model parameters, and fitting:

sherpa> PARAMPROMPT OFF
Model parameter prompting is off
sherpa> POLY[model1]
sherpa> model1 INTEGRATE OFF
sherpa> SOURCE = model1
sherpa> THAW model1.c1
sherpa> FIT
 LVMQT: V2.0
 LVMQT: initial statistic value = 2815.14
 LVMQT: final statistic value = 151.827 at iteration 5
        model1.c0  1.58227     
        model1.c1  0.198455

Then, these data and fit may be plotted:

sherpa> LPLOT FIT

Next, we thaw another parameter, fit again, and plot this fit:

sherpa> THAW model1.c2
sherpa> FIT
 LVMQT: V2.0
 LVMQT: initial statistic value = 151.827
 LVMQT: final statistic value = 59.0027 at iteration 4
        model1.c0  1.30826     
        model1.c1  0.347303     
        model1.c2  -0.0135317     

sherpa> LPLOT FIT

Figure 2 shows the resulting plot.

Modifying Plots

Plotting in Sherpa works in concert with the CIAO plotting tool ChIPS. ChIPS commands may be issued from within Sherpa to modify plots. For example, the following ChIPS commands may be issued to add labels to the X and Y axes of the existing plot:

sherpa> XLABEL "X = Off-Axis (arcmin)"
sherpa> YLABEL "F(X) = SNR"
sherpa> REDRAW

Figure 3 shows the resulting plot. Note that the command REDRAW must be issued in order for the Sherpa plot to be updated with previously issued ChIPS commands.

An important plotting concept is the numbering of curves. For example, in the current plot there are two curves; the data are curve number 1 and the red fit is curve number 2. To change the fit curve from histogram to a line, we issue the following command:

sherpa> C 2 SIMPLELINE
sherpa> REDRAW

Where, C 2 specifies that the SIMPLELINE command is to be applied to curve number 2. Similarly, we make a modification to curve number 1 (change the symbol style to a cross):

sherpa> C 1 SYMBOL CROSS
sherpa> REDRAW

The ChIPS command INFO is useful for getting information on the number of existing curves, etc:

sherpa> INFO

Drawing Area  #1  << CURRENT DRAWING AREA
(Location: 0.15 0.9 0.1 0.9) 
(Limits  : -7.45058e-09 11 1.37791 3.59135) 
(Axes    :  fouraxes   color: defaultcolor  width: 1) 
        Curve #1  
                 SYMBOL: defaultcolor  cross 
                 CONNECTOR: noline  << CURRENT CURVE
        Curve #2 
                 SYMBOL: none 
                 CONNECTOR: red  simpleline 
                 LINE: solid

We next make another plot modification using the ChIPS command ERRS, to change the style of the errorbars:

sherpa> ERRS STANDARD
sherpa> REDRAW

Figure 4 shows the resulting plot.

To add labels to the plot, we use the ChIPS command LABEL:

sherpa> LABEL 4.0 2.0 "F(X)=1.3083+0.3473X-0.0135X^2"

where the 4.0 2.0 coordinates specify the location on the plot for the placement of the label. To change this placement, the command is:

sherpa> L 1 4.0 1.5
sherpa> REDRAW

We may also add a second label:

sherpa> LABEL 1.0 3.3 "ACIS"

Now, to delete the first label:

sherpa> L 1 DEL
sherpa> REDRAW

To change the placement of the second label:

sherpa> L 2 4.0 2.0
sherpa> REDRAW

Note that, like curves, labels are also numbered. And, the L # and DEL commands must specify the label number.

Note that plots and their data may be saved: see the subsequent section Saving A Sherpa Session And Plot.

Plotting Into Multiple Windows

One may plot into multiple windows. For example, the following command plots the fit in one window and the fit residuals in another:

sherpa> LPLOT 2 FIT RESIDUALS

Figure 5 shows the resulting plot.

Various modifications may then be made to this plot using ChIPS commands:

sherpa>   # Change the data and fit plots in the 1st drawing
sherpa>   # area (D 1) to block symbols, and a line, respectively:
sherpa> D 1 C 1 NOLINE
sherpa> D 1 C 2 SIMPLELINE
sherpa> REDRAW

Note that drawing areas are also numbered, from top to bottom. In the above, we specify the drawing area number with D #. For example, D 1 C 1 NOLINE changes curve number 1 in drawing area number 1, and D 1 C 2 SIMPLELINE changes curve number 2 in drawing area number 1.

sherpa>   # Add labels to the X and Y Axes:
sherpa> D 2 XLABEL "X = Off-Axis (arcmin)"
sherpa> D 1 YLABEL "F(X) = SNR"
sherpa> REDRAW

Again, note that we specify the drawing area for the labels. For example, D 2 XLABEL adds an X axis label to drawing area number 2 (i.e. the bottom-most drawing area), and D 1 YLABEL adds a Y axis label to drawing area number 1.

sherpa>   # Add a title:
sherpa> TITLE "ACIS 25000 Counts Per Chip"
sherpa> REDRAW

sherpa>   # Remove the X Axis label from the 1st drawing area:
sherpa> D 1 XLABEL ""
sherpa> REDRAW

sherpa>   # Place a separation between the two drawing areas:
sherpa> SPLIT GAP y 0.04
sherpa> REDRAW

sherpa>   # Remove the X Axis tick marks from the 1st drawing area:
sherpa> D 1 TICKVALS X OFF
sherpa> REDRAW

sherpa>   # Change the format of the tick value labels on the Y Axes:
sherpa> D ALL TICKVALS y "%1.2f"
sherpa> REDRAW

Note that the "%1.2f" portion of the TICKVALS command specifies the format of the tick value labels using the standard C string format notation (see man printf).

sherpa>   # Add a label that contains the fit results:
sherpa> D 1 LABEL 2.0 1.7 "F(X)=1.3083+0.3473X-0.0135X^2"
sherpa> REDRAW

Note that text for labels may include TeX syntax for formatting text and creating mathematical expressions.

Figure 6 shows the resulting plot.

Printing Plots To PostScript

To create a hard copy of an existing plot, the command is PRINT:

sherpa> PRINT POSTFILE myplot1.ps

This command writes the plot to a PostScript file, which we here named myplot1.ps. This file may then be sent to the printer, perhaps as follows:

sherpa> $lpr myplot1.ps

Saving A Sherpa Session And Plot

It is important to note that the data displayed in the current plot and the commands used to generate the plot are not written to disk by default. To save the current Sherpa plot, and its associated data:

The plot and its data may be saved using the following ChIPS command:
```
sherpa> STORE myplot1.chp
```
This command specifies that a record of the commands used to generate the plot be saved in a file named myplot1.chp. The STORE command also creates a FITS file that contains the plotted data (e.g. myplot1.chp.fits).

We can confirm that the myplot1.chp file and the myplot1.chp.fits datafile were indeed written:
```
sherpa> ls myplot1*
myplot1.chp             myplot1.chp.fits
```
Examine the myplot1.chp using the Unix command more and note that this file contains all of the ChIPS commands necessary to create the plot; it may be edited to further modify the plot.
```
sherpa> $more myplot1.chp
#This is a ChIPS state file.
#ChIPS_Version: 2.20


redraw off


##############################
# Start of a new drawing area.
##############################
drawarea 0.15 0.9 0.52 0.9
label 2 1.7 "F(X)=1.3083+0.3473X-0.0135X^2"

#######################
# Start of a new curve.
skip 0
curve "/export/work/sherpa_threads/sherpa/plot/myplot1.chp.new.fits[curve5]" yup 3 ydn 4 
symbol  square 
symbol 4 0
symbol 2
errs x size 2
.
.
.
```
In addition to saving the plot and the data used in the plot, we may also save the current Sherpa session:
```
sherpa> SAVE ALL mysession1.shp
```

Restoring A Sherpa Plot And Session

We may now exit Sherpa:

sherpa> EXIT
Goodbye.

And, later return to where we left off:

unix% sherpa
sherpa> RESTORE myplot1.chp

The above RESTORE command restores the plot. Note that the plot is restored using the commands recorded in the myplot1.chp file, and the myplot1.chp.fits datafile that was output and is referenced in the myplot1.chp file.

We are now able to make additional changes to the plot:

sherpa> TITLE BLUE
sherpa> REDRAW

It is important to note, however, that the Sherpa session has not yet been restored:

sherpa> SHOW

Optimization Method: Levenberg-Marquardt
Statistic:           Chi-Squared Gehrels

To restore the Sherpa session:

sherpa> USE mysession1.shp

sherpa> SHOW

Optimization Method: Levenberg-Marquardt
Statistic:           Chi-Squared Gehrels

-----------------
Input data files:
-----------------

Data 1: data1.dat ascii 1 2.
Total Size: 11 bins (or pixels)

(etc)

We are now able to continue with the fitting session:

sherpa> THAW model1.c3
sherpa> FIT
 LVMQT: V2.0
 LVMQT: initial statistic value = 59.0027
 LVMQT: final statistic value = 30.8491 at iteration 6
        model1.c0  1.49843     
        model1.c1  0.1447     
        model1.c2  0.0322936     
        model1.c3  -0.00277729     

sherpa> LPLOT 2 FIT RESIDUALS

Note that a plot saved from within Sherpa may also be restored into ChIPS:

unix% chips

Welcome to ChIPS, version CIAO 3.4
Copyright (C) 1999-2003, Smithsonian Astrophysical Observatory

chips> RESTORE myplot1.chp

It is sometimes more convenient to work within ChIPS itself when performing further plot modifications.

Saving Plotted Data To Output Files

As before, we can improve this plot using ChIPS commands:

sherpa> LPLOT 2 FIT RESIDUALS
sherpa> D 1 C 1 NOLINE
sherpa> D 1 C 2 SIMPLELINE
sherpa> D 2 XLABEL "X = Off-Axis (arcmin)"
sherpa> D 1 YLABEL "F(X) = SNR"
sherpa> TITLE "ACIS 25000 Counts Per Chip"
sherpa> D 1 XLABEL ""
sherpa> SPLIT GAP y 0.04
sherpa> D 1 TICKVALS X OFF
sherpa> D 1 TICKVALS y "%1.2f"
sherpa> D 2 TICKVALS y "%1.2f"
sherpa> D 1 LABEL 2.0 1.7 "F(X)=1.4984+0.1447X+0.0323X^2-0.0028X^3"
sherpa> REDRAW

And also as before, we can save the data being plotted, including this new fit, using the STORE command:

sherpa> STORE myplot2.chp

This command writes the plotted data to the current directory, and names the file myplot2.chp.fits. The commands used to generate the plot are saved to a file named myplot2.chp.

If we'd also like to save the Sherpa session, to possibly return to later, then we must issue the SAVE ALL command:

sherpa> SAVE ALL mysession2.shp
sherpa> EXIT
Goodbye.

Restoring Only A Plot

A plot may be restored either in ChIPS, or in Sherpa:

To restore a plot in ChIPS:
```
unix% chips
chips> RESTORE myplot2.chp
```
One may then issue further ChIPS commands to modify the plot.
To restore a plot in Sherpa:
```
unix% sherpa
sherpa> RESTORE myplot2.chp
```
Figure 7 shows the resulting plot. One may then issue further ChIPS commands to modify the plot.

In addition to restoring the plot, we may also restore the previously saved Sherpa session:
```
sherpa> USE mysession2.shp
.
.
sherpa> EXIT
Goodbye.
```

Overplotting

Two different datasets may be plotted separately in different drawing areas:

unix% sherpa
sherpa> DATA 1 data1.dat
sherpa> DATA 2 data2.dat
sherpa> LPLOT 2 DATA 1 DATA 2
sherpa> D 1 C 1 ERRS NONE
sherpa> D 1 C 1 SYMBOL NONE
sherpa> D 1 C 1 HISTO
sherpa> D 2 C 1 ERRS NONE
sherpa> D 2 C 1 SYMBOL NONE
sherpa> D 2 C 1 HISTO
sherpa> REDRAW

Figure 8 shows the resulting plot.

Or, one may overplot two different datasets in the same drawing area, using the Sherpa OPLOT command:

sherpa> CLEAR
sherpa> OPLOT DATA 1 DATA 2

Figure 9 shows the resulting plot.

Note that overplotting effectively works only for datasets having the same data space units (i.e. energy, wavelength, PHA bin, etc.).

Saving Spectral Data In Energy Space

Without an instrument model definition, input PHA data are in count space. For example,

sherpa> ERASE ALL
sherpa> DATA data3.pha
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: using systematic errors specified in the PHA file.
sherpa> LPLOT DATA

Figure 10 shows the resulting plot. To save these data to an output file, one may either use the Sherpa WRITE command, or one may save the data from the plot. Here, we save the data from the plot:

sherpa> STORE myplot3_channel.chp

This will result in a myplot3_channel.chp file that contains the commands needed to regenerate the plot, and a myplot3_channel.chp.fits datafile that contains the data (in ASCII).

To convert the PHA data to energy space, one must define an instrument model. For example,

sherpa> RSP[inst3](data3_rmf.fits, data3_arf.fits)
The inferred file type is ARF.  If this is not what you want, please 
specify the type explicitly in the data command.
sherpa> INSTRUMENT = inst3
sherpa> LPLOT DATA

Figure 11 shows the resulting plot. To save these data (which are now in energy space) to an output file, one must save the data from the plot:

sherpa> STORE myplot3_energy.chp

This will result in a myplot3_energy.chp file that contains the commands needed to regenerate the plot, and a myplot3_energy.chp.fits datafile that contains the data in energy space (in ASCII).

One may EXIT, and then later regenerate these plots:

sherpa> EXIT
Goodbye.

unix% sherpa
sherpa> RESTORE myplot3_channel.chp
sherpa> RESTORE myplot3_energy.chp

Note that these plots may be regenerated in either Sherpa or ChIPS, and that these plots make use of the data saved in the myplot3_channel.chp.fits and myplot3_energy.chp.fits datafiles.

We may now exit Sherpa:

sherpa> EXIT
Goodbye.

Overplotting Two Spectra In Energy Space

In order to overplot two spectra in energy space, one must first individually plot each spectrum in energy space. These plotted data are saved and are then overplotted using ChIPS. The procedure is as follows:

First, input the two spectra:

unix% sherpa
sherpa> DATA 1 srcc.pi
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ ERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin
RMF is being input from:
  /export/work/sherpa_threads/sherpa/plot/srcc.rmf
ARF is being input from:
  /export/work/sherpa_threads/sherpa/plot/srcc.arf
Background data are being input from:
  /export/work/sherpa_threads/sherpa/plot/srcc_bg.pi
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ BERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin

sherpa> IGNORE 1 BAD

sherpa> DATA 2 srce.pi
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ ERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin
RMF is being input from:
  /export/work/sherpa_threads/sherpa/plot/srce.rmf
ARF is being input from:
  /export/work/sherpa_threads/sherpa/plot/srce.arf
Background data are being input from:
  /export/work/sherpa_threads/sherpa/plot/srce_bg.pi
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ BERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin

sherpa> IGNORE 2 BAD

Note that RMF and ARF files were automatically loaded since the input datafiles contained headers containing paths to these calibration files. As such, instrument models are automatically defined by Sherpa for these datasets. So when the data are plotted, they are plotted in energy space:

sherpa> LPLOT 2 DATA 1 DATA 2

These energy space data may also be overplotted, using the Sherpa OPLOT command:

sherpa> OPLOT DATA 1 DATA 2
sherpa> C 1 NOLINE
sherpa> C 2 NOLINE
sherpa> C 1 ERRS BOTH
sherpa> C 2 ERRS BOTH
sherpa> C 1 SYMBOL SQUARE
sherpa> C 1 SYMBOL SIZE 2.0
sherpa> C 1 SYMBOL RED
sherpa> C 2 SYMBOL SQUARE
sherpa> C 2 SYMBOL SIZE 2.0
sherpa> REDRAW

Figure 12 shows the resulting plot.

Overplotting Model Components

The following is an example of fitting and then overplotting the model components:

sherpa> ERASE ALL
sherpa> DATA data5_pi.fits
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ ERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin
sherpa> RSP[detector1](data5_rmf.fits, data5_arf.fits)
The inferred file type is ARF.  If this is not what you want, please 
specify the type explicitly in the data command.
sherpa> INSTRUMENT = detector1
sherpa> IGNORE BAD
sherpa> METHOD POWELL

sherpa> PARAMPROMPT OFF
Model parameter prompting is off
sherpa> SOURCE = XSPHABS[abs] * POWLAW1D[power]
sherpa> abs.nH.value = 0.0454
sherpa> FREEZE abs.nH
sherpa> FIT
 powll: v1.2
 powll:  initial statistic value =     2.38800E+03
 powll:     converged to minimum =     2.23713E+02 at iteration =      4
 powll:    final statistic value =     2.23713E+02
         power.gamma  1.87995     
         power.ampl  0.000371676     

sherpa> GAUSS1D[g]
sherpa> g.pos=6.0
sherpa> g.pos.min=5.0
sherpa> g.pos.max=7.0
sherpa> SOURCE = abs*power + g
sherpa> FIT
 powll: v1.2
 powll:  initial statistic value =     8.71340E+05
 powll:     converged to minimum =     2.22480E+02 at iteration =     21
 powll:    final statistic value =     2.22480E+02
         power.gamma  1.88101     
         power.ampl  0.000371562     
             g.fwhm  0.104683     
             g.pos  6.82238     
             g.ampl  3.28794e-05     

sherpa> GAUSS1D[g2]
sherpa> SOURCE = abs*power + g + g2
sherpa> g2.pos=6.5
sherpa> g2.pos.min=6.0
sherpa> g2.pos.max=7.0
sherpa> FIT
 powll: v1.2
 powll:  initial statistic value =     6.51744E+05
 powll:     converged to minimum =     2.22464E+02 at iteration =     20
 powll:    final statistic value =     2.22464E+02
         power.gamma  1.881     
         power.ampl  0.000371562     
             g.fwhm  0.0386909     
             g.pos  6.76242     
             g.ampl  2.79281e-05     
            g2.fwhm  0.0228727     
            g2.pos  6.84466     
            g2.ampl  0.000103375     

WARNING:
   The value of g2.fwhm is equal to the g2.fwhm.min limit boundary.
   You may wish to consider changing min/max values and refitting.

sherpa> LPLOT FIT
sherpa> OPLOT SOURCE g g2

Figure 13 shows the resulting plot.

2-D Data

Introduction to 2-D Data Display in Sherpa

2-D data may be visualized with either a contour plot or a surface plot, using the plotting commands CPLOT and SPLOT respectively. For 2-D imaging data, the command is IMAGE. In addition, the REGION-PROJECTION command creates a contour plot of confidence regions, computed using the PROJECTION algorithm, and the REGION-UNCERTAINTY command creates a contour plot of confidence regions, computed using the UNCERTAINTY algorithm. Parts of this thread demonstrate 2-D data visualization using REGION-PROJECTION in detail.

2-D Plotting, Contour Plots

Here, 2-D data is input and then displayed using a contour plot:

sherpa> ERASE ALL
sherpa> READ DATA data6.dat ASCII 1 2 3
sherpa> CPLOT DATA
Contour Levels: 598 448.637 299.274 149.911 0.547731 
Min: 0, Max: 598, Ave: 0.547731

Figure 14 shows the resulting plot.

Note that the plot levels may be modified using the ChIPS command LEVELS:

sherpa> LEVELS 3 5 20 100
sherpa> REDRAW

2-D Plotting, Surface Plots

These data may also be displayed using a surface plot:

sherpa> SPLOT DATA

Figure 15 shows the resulting plot.

Here, a second 2-D dataset is input, and then the two datasets are displayed into multiple windows, using surface plots:

sherpa> READ DATA 2 data7.dat ASCII 1 2 3
sherpa> SPLOT 2 DATA 1 DATA 2

Figure 16 shows the resulting plot.

2-D Plotting, REGION-PROJECTION

The current session is erased before inputting new data:

sherpa> ERASE ALL

sherpa> READ DATA data4_pi.fits
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ ERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin

These 1-D data may then be plotted from within Sherpa:

sherpa> LPLOT DATA

Next, we continue the Sherpa session by defining an instrument model, displaying the data in energy space, and filtering the dataset:

sherpa> PARAMPROMPT OFF
Model parameter prompting is off
sherpa> RSP[acis](data4_rmf.fits, data4_arf.fits)
The inferred file type is ARF.  If this is not what you want, please 
specify the type explicitly in the data command.
sherpa> INSTRUMENT = acis
sherpa> LPLOT DATA 
sherpa> IGNORE ENERGY :0.5, 8:
sherpa> LPLOT DATA

Next, model components and a source model are defined, the statistic is changed, and then the data are fit:

sherpa> POWLAW1D[p1]
sherpa> XSWABS[abs1]
sherpa> SOURCE = abs1*p1
sherpa> GUESS SOURCE 
sherpa> STATISTIC CASH
sherpa> METHOD POWELL

sherpa> FIT 
 powll: v1.2
 powll:  initial statistic value =     1.53063E+04
 powll:     converged to minimum =    -7.62300E+03 at iteration =     12
 powll:    final statistic value =    -7.62300E+03
          abs1.nH  1.48605  10^22/cm^2  
            p1.gamma  0.815361     
            p1.ampl  0.000711629     

WARNING:
   The value of p1.ampl within 0.01% of the p1.ampl.max limit boundary.
   You may wish to consider changing min/max values and refitting.

sherpa> p1.ampl.max = 0.1
sherpa> FIT
 powll: v1.2
 powll:  initial statistic value =    -7.62300E+03
 powll:     converged to minimum =    -7.70549E+03 at iteration =      7
 powll:    final statistic value =    -7.70549E+03
          abs1.nH  2.37765  10^22/cm^2  
            p1.gamma  1.50086     
            p1.ampl  0.00200489     

sherpa> LPLOT FIT
  ==> Error bars computed using Chi Gehrels.

Figure 17 shows the resulting plot of the fit.

Finally, we next use the REGION-PROJECTION command to create a 2-D contour plot of confidence regions, computed using the PROJECTION algorithm:

sherpa> REG-PROJ abs1.nH p1.gamma
Region-Projection: optimization reset to Simplex.
                   computing grid size with covariance...done.
                   outer grid loop 20% done...
                   outer grid loop 40% done...
                   outer grid loop 60% done...
                   outer grid loop 80% done...
Minimum: -7705.49
Levels are: -7703.2 -7699.31 -7693.66
-----------------------------------------------------------------
NOTE: the limits on this plot cannot be altered.
      * To change the overall box size, you can:
        a) Set sherpa.regproj.expfac to a new value
           (if sherpa.regproj.arange = 1); or
        b) Set sherpa.regproj.min and max to new values
           (if sherpa.regproj.arange = 0).
      * To create a denser grid, set sherpa.regproj.nloop
        to new values.
      * To see current settings, type list_regproj().
      * For more information, do AHELP REGION-PROJECTION.
      This message will not be printed again during this session.
-----------------------------------------------------------------

The REGION-PROJECTION parameter values are then modified as follows, and REGION-PROJECTION is then run again to produce another 2-D contour plot of the confidence regions:

sherpa> sherpa.regproj.arange = 0
sherpa> sherpa.regproj.min = [1.90,1.20]
sherpa> sherpa.regproj.max = [2.85,1.80]
sherpa> sherpa.regproj.nloop = [30,30]
sherpa> REG-PROJ abs1.nH p1.gamma
Region-Projection: optimization reset to Simplex.
                   grid size set by user.
                   outer grid loop 20% done...
                   outer grid loop 40% done...
                   outer grid loop 60% done...
                   outer grid loop 80% done...
Minimum: -7705.49
Levels are: -7703.2 -7699.31 -7693.66

Figure 18 shows the resulting contour plot.

Modifying Plots

As with 1-D plotting, 2-D plotting in Sherpa also works in concert with the CIAO plotting tool ChIPS. So, ChIPS commands may be issued from within Sherpa to modify 2-D plots. For example, the following ChIPS commands may be issued to modify the existing contour plot:

sherpa> TWOAXES
sherpa> RED
sherpa> WIDTH 4.0
sherpa> REDRAW

Note that the command REDRAW must be issued in order for the plot to be updated with previously issued ChIPS commands.

Note, however, that axis limits on contour and surface plots may not be modified.

Printing Plots To PostScript

Just as with 1-D plots, to create a hard copy of an existing 2-D plot the command is PRINT:

sherpa> PRINT POSTFILE myplot4.ps

This command writes the plot to a PostScript file, which we here named myplot4.ps. This file may then be sent to the printer, perhaps as follows:

sherpa> $lpr myplot4.ps

Saving A Sherpa Session, Plot, and Plotted Data

Just as with 1-D plots, the data displayed in the current plot and the commands used to generate the plot are not written to disk by default. To save the current Sherpa plot, and its associated data:

The plot may be saved using the following command:
```
sherpa> STORE myplot4.chp
```
This command specifies that a record of the commands used to generate the plot be saved in a file named myplot4.chp. The dataset used in the plot will be saved and named /tmp/<pid>sherpa_cont_params.ascii, where <pid> is a number that is related to the process ID of the current Sherpa session.

We can confirm that the myplot4.chp file and the /tmp/<pid>sherpa_cont_params.ascii datafiles were indeed written:
```
sherpa> ls myplot4*
myplot4.chp
sherpa> ls /tmp/*sherpa_cont_params.ascii
/tmp/26978sherpa_cont_params.ascii
```
We see that there is one datafile, as expected: /tmp/26978sherpa_cont_params.ascii contains the data being plotted. The contents of the myplot4.chp file are the commands needed to regenerate the plot, including commands to re-load this datafile:
```
sherpa> $more myplot4.chp
.
.
contour "/tmp/26978sherpa_cont_params.ascii"
```
Since the datafile in /tmp/ will be deleted when the Sherpa session ends, it should be copied to the local directory:
```
sherpa> $cp /tmp/26978sherpa_cont_params.ascii myplot4.chp.dat
```
In addition to saving the plot and the data used in the plot, we may also save the current Sherpa session:
```
sherpa> SAVE ALL mysession4.shp
```
We may now exit Sherpa:
```
sherpa> EXIT
Goodbye.
```
And, later return to where we left off.
The user will then need to edit the myplot4.chp file so that it points to the datafile myplot4.chp.dat, instead of to /tmp/26978sherpa_cont_params.ascii.

Restoring A Sherpa Plot And Session

unix% sherpa
sherpa> RESTORE myplot4.chp

The above RESTORE command restores the plot. Note that the plot is restored using the commands recorded in the myplot4.chp file, and the myplot4.chp.dat datafile that was output and is referenced in the myplot4.chp file.

We are now able to make additional changes to the plot:

sherpa> TITLE BLUE
sherpa> REDRAW

It is important to note, however, that the Sherpa session has not yet been restored.

To restore the Sherpa session and continue with further analyses:

sherpa> USE mysession4.shp
The inferred file type is ARF.  If this is not what you want, please 
specify the type explicitly in the data command.
The inferred file type is PHA.  If this is not what you want, please 
specify the type explicitly in the data command.
WARNING: statistical errors specified in the PHA file.
         These are currently IGNORED.  To use them, type:
         READ ERRORS "<filename>[cols CHANNEL,STAT_ERR]" fitsbin
sherpa> SHOW
.
.
.

History

14 Jan 2005	reviewed for CIAO 3.2: no changes
21 Dec 2005	reviewed for CIAO 3.3: no changes
01 Dec 2006	updated for CIAO 3.4: ChIPS version

Data Visualization

Overview

Contents

Getting Started

1-D Data

Introduction to 1-D Data Display in Sherpa

1-D Plotting

Modifying Plots

Plotting Into Multiple Windows

Printing Plots To PostScript

Saving A Sherpa Session And Plot

Restoring A Sherpa Plot And Session

Saving Plotted Data To Output Files

Restoring Only A Plot

Overplotting

Saving Spectral Data In Energy Space

Overplotting Two Spectra In Energy Space

Overplotting Model Components

2-D Data

Introduction to 2-D Data Display in Sherpa

2-D Plotting, Contour Plots

2-D Plotting, Surface Plots

2-D Plotting, REGION-PROJECTION

Modifying Plots

Printing Plots To PostScript

Saving A Sherpa Session, Plot, and Plotted Data

Restoring A Sherpa Plot And Session

History