Hello I was wondering if anyone had an old working batch file that I can look at I'm hoping to use it as reference in the proper construction of my own.

Thanks a lot

Stephanos Mitchell Henein B.Sc, B.A
Agriculture and Agri-Food Canada/Agriculture et Agroalimentaire Canada
GIS Analyst & Support
Telephone/Téléphone: 306-780-6567
Facsimile/Télécopieur: 306-780-5018
1800 Hamilton Street, Suite 408
Regina, Saskatchewan
S4P4L2
[log in to unmask]

 Agriculture and Agri-Food Canada - Agriculture et Agroalimentaire Canada

-----Original Message-----
From: DSSAT - Crop Models and Applications [mailto:[log in to unmask]] On Behalf Of Jeffrey White
Sent: Wednesday, March 07, 2007 4:22 PM
To: [log in to unmask]
Subject: Batch files

Steve Henein inquired about experiences running DSSAT with batch files. I
am currently working with a large set (~400) of winter wheat experiments
and do much of the simulations with batch files.



Some notes from my work:

1. Once I had the full set of File-Xs, I wrote a Python script to compare
cultivar names and treatment names. (I was concerned that increating the
File-Xs that errors might have crept in.) The output file listed the
treatment numbers and file names, or you could get the whole list by
letting DSSAT create a batch file for your entire experiment list (you
wouldn't even have to run it -- just create the batch file).

2. I found that you can write annotations to the right of TRNO and the
batch file runs fine, so I try to add cultivar names or locations.

3. Here's the neat part... From UEdit, my favorite text editor, you can
launch batch files by hitting F10 and providing the command info. For
example: C:\DSSAT4\DSCSM045.EXE t d4batch.dv4 This allows me to run CSM
very quickly after editing a file, such as in calibrating cultivars or
checking a new file-X.

4. I keep a large set of batch files for the wheat data. One runs all
experiments. Another pair run calibration and validation data subsets. A
large series corresponds to calibration subsets for individual cultivars.
Thus, if a cultivar appears in 20 treatment x experiment combinations, I
have reserved 7 or 8 for calibration.



In the last DSSAT training workshop we discussed whether we need a power
users meeting to look at these sorts of approaches. Something else I do
that works pretty well is create FileXs using SAS. There's no way I'd
create 400 file-Xs by hand editing. Unfortunately, SAS is too expensive
for many users, so I am constantly looking for better tools for
manipulating large sets of data. The stat package R looks excellent for
regressions, plotting, and an incredible array of advanced stat methods,
but it does not provide the kinds of sort, merge and formatted output that
SAS does very well.



Best regards,

Jeff White



Jeffrey W. White US Arid Land

Agric Research Center, USDA-ARS

21881 North Cardon Lane

Maricopa, AZ 85239 USA

Tel. 1-520-316-6368  ::  Fax 1-520-316-6330