isis-0.9.8 installed

Changes:

* I fixed $PWD glitch reported by Dave H

* In response to comments from Dave D. and John D., I changed 
  the fit/eval/renorm interface *again*. The new function names are:
  
     fit_counts, fit_flux
     eval_counts, eval_flux
     renorm_counts, renorm_flux
     
  See the help pages for fit_* for details.     

* Based on Dan's suggestions, I revised the fit behavior
  to eliminate the need to unassign_arf/unassign_rmf 
  in order to apply an ideal instrument response (equivalent
  to not applying the response) when arfs and rmfs have
  already been assigned.
  
  See the help pages for fit_* for details.


At this point, the fitting routines could use some additional
testing, considering of the large number of modes available:

fit_counts;
flux_corr, fit_flux;

assign_arf, assign_rmf, fit_counts
assign_arf, assign_rmf, fit_counts(0);
assign_arf, assign_rmf, flux_corr, fit_flux

assign_arf, assign_rmf, xnotice, fit_counts
assign_arf, assign_rmf, xnotice, fit_counts(0);
assign_arf, assign_rmf, xnotice, flux_corr, fit_flux

assign_arf, assign_rmf, group_data, fit_counts
assign_arf, assign_rmf, group_data, fit_counts(0);
assign_arf, assign_rmf, group_data, flux_corr, fit_flux

assign_arf, assign_rmf, group_data, xnotice, fit_counts
assign_arf, assign_rmf, group_data, xnotice, fit_counts(0);
assign_arf, assign_rmf, group_data, xnotice, flux_corr, fit_flux

It would also be good to test for the case when the RMF is
non-square and the ARF has more (or fewer) bins than the data,
especially the interaction of this with grouping and with
ignore/notice.

And then we can repeat relevant tests using the pileup kernel.

In fact, it would be really nice to have a regression script to
test all of these modes automatically, with suitably chosen test
data....

Or perhaps someone would like to suggest that I add a few more
fitting modes?  (simultaneously fitting background is already on
the list).

John


  
----
You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
unsubscribe