Use slycot's tb05ad for faster and more accurate frequency response e… #172
Conversation
Added IO cleanup function to xferfcn module.
From .eggs/README.txt: "This directory contains eggs that were downloaded by setuptools to build, test, and run plug-ins. This directory caches those eggs to prevent repeated downloads. However, it is safe to delete this directory."
…function. Added unit tests to test the cleanup function. Added deep copy to TransferFunction args.
…n-control into jed-frey-InputSanitization
…tion Updates to input sanitization
…valuation of state-space systems. If slycot is unavailible, use the built in horner function (instead of converting to a transfer function, as was done before).
Travis CI update for new/old scipy + improved slycot
Update conda-recipe for use with conda-build >= 3.0
Fixes for SciPy 1.0 compatibility
Cleanup TransferFunction Input Sanitization
…or_doc Update docstrings for LTI class constructors
|
@rabraker Can you rebase these changes against the current master. That should fix up the problems in Travis CI and allow testing against python 3.5 and 3.6. |
Fix minreal not returning a discrete TF
…valuation of state-space systems. If slycot is unavailable, use the built in horner function (instead of converting to a transfer function, as was done before).
…into freqresp_tb05ad
|
Changes Unknown when pulling 308d8b5 on rabraker:freqresp_tb05ad into ** on python-control:master**. |
|
I'm evidently all thumbs with this rebasing business. Whatever I did makes all the checks pass, but somehow there are commits from others going back almost a year in this PR now. I don't know how to fix that. So it seems easiest to me to delete this PR and I'll just make a new, clean one. |
|
Pretty sure there is a way to rewrite history that will make this a clean commit. @slivingston is a master at git, perhaps he has some suggestions? One possibility is that you didn't grab the upstream changes from python-control:master and that is causing the large number of extraneous commits. As I check, I did a manual merge with my copy of the master branch and the only change was to statesp.py. I think if you do |
|
With my branch ie, this had no effect. I'll just close this PR and submit another. |
This pull request addresses issue #116.
It uses the (relatively) newly wrapped TB05AD function from Slycot, when Slycot is available. If slycot is not available, I instead use the state space built-in method
horner, rather than converting to a transfer function for the evaluation.This passes the unit tests on my machine (python 2.7), but currently it looks like the travis build is failing so I'm not sure what happens with other versions.