Activity for Rob Clewley

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Maurizio, thanks for your continued interest! Clewley R, Hybrid Models and Biological Model Reduction with PyDSTool, PLoS Computational Biology 8(8): e1002628 (2012) Clewley RH, Sherwood WE, LaMar MD, Guckenheimer JM (2007) PyDSTool, a software environment for dynamical systems modeling. URL http://pydstool.sourceforge.net This info is still available from the Google cache of the original front page and from the waybackmachine's version, https://web.archive.org/web/2/http://www.ni.gsu.edu/~rclewley/pydstool/FrontPage.html...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I will try to re-host the pages on my github account in the near future. Sorry for the unexpected outage. In the meantime, this should be functional, if a bit slow: https://web.archive.org/web/20160331125637/http://www.ni.gsu.edu/~rclewley/PyDSTool/ On Mon, Dec 18, 2017 at 6:16 AM, Maurizio De Pitta' mauriziodepitta@users.sf.net wrote: Uhm ok, there seems to be a renovation going on as I read on the current Fron Page at http://www2.gsu.edu/~matrhc/FrontPage.html Site down Sent from sourceforge.net...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    This is definitely a problem, thanks for posting. I've had some success with addressing it here (https://github.com/robclewley/pydstool/pull/128) but it's not complete and I don't know how to proceed right now. Dopri is definitely the superior way in all cases, though ...

  • Rob Clewley Rob Clewley posted a comment on discussion Open Discussion

    Hi. Good question! It has been a long time since that code was written and the full change logs weren't even in git at that point. I recall hacking together some quick min and max helper functions in C but had trouble (or lack of time) making fully generic, i.e. arbitrary number of arguments. Somehow it ended up with 4 as the limit. I can't even see the evidence of where 4 is a constraint in the C definitions for the __maxof and __minof underlying functions, so it's possible they were updated by...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Well, you can't unload and reload the same named module (i.e. the DLL created from the C code). So, you have to at least a whole new python process to be able to make that visible after delete/recreate. At the python level, merely recreating the generator is not sufficient to do a deep-level restart of the DLL and its memory allocation. You can time-domain split based on state-dependent events just as easily as by literal time. Just spawn subprocesses that run the next part until either a time limit...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Sorry for the difficulty with this. cleanupMemory was our early attempt to mitigate the problem but I agree it doesn't seem to have helped. You definitely can't recreate the generator object -- the DLL will remain the same under the hood. I simply meant that, in my experience, there are fewer segfaults if one splits the time domain into chunks and compute trajectories over those restricted domains consecutively. That way, the same blocks of memory will be overwritten with each new call to compute,...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, Ahmet could you paste the specific error messages you received for your two approaches to designing your RHS? On Thu, May 18, 2017 at 9:42 AM, Drew LaMar mdlamar@users.sf.net wrote: Actually, just realized this is not PyCont related, so it's probably best if another member of the team chases after this one. Also just noticed that you are creating things symbolically, which I don't think VFGEN deals with. Anyone else from the PyDSTool team wanna try their hand at this? corticotroph cell model...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    My understand is that your compiler version has to match that used to build scipy within Anaconda. Your Windows version of Anaconda seems to have been built with the microsoft C compiler, which is proprietary. For creating your own C-based extensions in the scipy stack, you are generally better off building your own system from source with the same compiler. I'm not sure what the current state of play is regarding fixing this from a prebuilt Anaconda, install, but follow these links and see if anything...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    The link doesn't end with a period, maybe it got mis-formatted somewhere. Let me know if you have more questions after you've taken a look at that page and the examples. -Rob On Fri, May 5, 2017 at 11:44 PM, Amardeep Mishra mishraiitm@users.sf.net wrote: Hi! Rob! Thanks a lot for replying Rob and Sorry for replying late as I was out of the town travelling! I couldnt open this link- http://www.ni.gsu.edu/~rclewley/PyDSTool/ParamEst.html. "Alternatively, you can take a much easier route: set up your...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Amardeep, Sure you can. The built-in parameter estimation tools are designed for exploration of highly non-convex, possibly poorly conditioned systems. They provide substantial diagnostic feedback and user supervision of steps / adjustments to algorithmic parameters. As such, it's not necessarily very fast. Check out the examples marked "pest_*" in the examples directory included and read http://www.ni.gsu.edu/~rclewley/PyDSTool/ParamEst.html. Alternatively, you can take a much easier route: set...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    The diagnostics returned with that error suggest that it's struggling to compute steps, and possibly the step size is shrinking to zero. I suspect it might be because you have to be more careful when adding a noisy input to what's meant to be a stationary, deterministic RHS definition! The deterministic ODE solver sees this "input" as a parameter that's changing, which can cause a lack of convergence if the amplitude is large enough or the frequency of change is high enough. Can you try running this...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Michel, yes the second approach is the only one available. You'll have to do some...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I see, I wasn't thinking about that. The system I described is set up for temporal,...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Absolutely, although the accuracy of manifold calcs will naturally depend on the...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Sorry this easy system has proven difficult to work. I can verify that I can get...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Thank you for the note. Looks like you fixed it properly. I'll roll it in to the...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Directly reusing legacy codes is not the approach that this software takes, but you...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    As the docs point out, both Dopri and Radau originate from Hairer and Wanner. While...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, PyDSTool's optimization tools are experimental and more suitable for initial...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    All the "pest" named examples have information about that capability. I'm not aware...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    That's fine but just bear in mind that many continuations take too long to be close...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You are correct, unfortunately. I am out of touch with recent developments techniques...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    No, unfortunately, having progress bars incur a lot of overhead, given that the integration...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    This is a python error rather than a PyDSTool error. You have to keep track of what...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    The line defining 'Mdm2nuc' uses Vratio as if it was a function. On Tue, Nov 8, 2016...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, if you email me your entire script I can see if I can understand what the issue...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    http://www.ni.gsu.edu/~rclewley/PyDSTool/FunctionalSpec.html#Spec shows a definition...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Your function definition should use the PyDSTool auxiliary function syntax. It's...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Your function definition should use the PyDSTool auxiliary function syntax. It's...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Well, you still have the empty string in the params for the event. That's not valid...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I just noticed your list of an empty string for parnames. Also, be careful to separate...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You can't have your special variable in "ignorespec" also be a variable name. Variables...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Off the top of my head, I'd say you have one primary option in the absence of modulation...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    If you capture the gcc command to console that build_ext (setuptools/distutils, in...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Yes, that's the right idea. You can try changing std but I don't know what will happen....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I assume this is related to an issue mentioned recently in another forum post. The...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    If your step size is tiny and your num_points is small, you may have to repeat calls...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    To add to Drew's points: If you plan to do continuation and bifurcation analysis,...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    It's all here, in the controls for time step: http://www.ni.gsu.edu/~rclewley/PyDSTool/Generators.html...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You found a bug in the parsing of embedded if statements for C-based integrators....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Can you confirm that dill will serialize the PyCont classes? Because I didn't think...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You could allow the data to be linearly interpolated, if you follow a scheme like...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Maybe what's best here, given our lack of resources, is for Danny to submit a PR...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Your example does not involve a dynamical system. What example are you trying to...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I don't have any references, sorry. I don't recall who has told me about that and...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    In lieu of any other responses, I'm not familiar with examples over 1000 personally....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Fabio, that is normal. The VODE wrapper is esesentially a kludge to provide some...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I'm not sure it's true that all the data has not been included, but I'm not familiar...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    They are all tolerances in tests for convergence of various algorithms. VarTol is...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi. Unfortunately, this is a long-standing issue that we are aware of and that has...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Cool. I wish there was an easier way to help the user catch those instances. Anyway,...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    ...dop853_temp//RH_J_vf.c:116: error: conflicting types for 'sinf' .../dop853_temp/RH_J_vf.c:330:...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I'm posting again because I don't think my previous reply went through properly (SF...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    We are not, but that's an interesting idea. It couldn't happen soon, in any case....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    SF emails are just username@users.sourceforge.net, so mine is robclewley@users.s...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Follow the provided examples that use fnspecs. You still have to follow the same...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Have you tried adjusting the matplotlib backend, adding a call to matplotlib's show...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    The best way is to have aux functions in fnspecs that take no arguments and which...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    There could be several reasons why the results are not the same. Can you be more...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I believe you can contact him directly from here: http://www.people.wm.edu/~mdla...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I'd like to help, but this isn't the part of the code that I developed and I'm not...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, This is Drew's code, but I think I can answer some of this, at least. Yes, you're...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You need to set auxvars to a list of names to set them to be simple algebraic definitions....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    In answer to your second question: yes, initcond is what you need.

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Glad to see you got the basics of a hybrid system up and running quickly! Your switching...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I see. That's not directly supported in a single Generator. Instead, you'll have...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You can simply multiply the event definition RHS's together. Since you're defining...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Although I'd like to get confirmation from Drew, who hasn't been around recently,...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I don't think there has been any progress on the full rewrite of the interfaces to...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I'm not sure I understand your question, but you probably mean the trajectory's getEvents...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Just use a list.

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, There is a bib provided in PyDSTool/PyCont/biblio_for_examples.txt. That covers...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Also, if this is pure L-V, then there are infinite fixed points because it's not...

  • Rob Clewley Rob Clewley posted a comment on discussion Open Discussion

    The INCF (http://incf.org) is sponsoring two Google Summer of Code projects this...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    The fixed point and nullcline algorithms are really just prototypes. They generally...

  • Rob Clewley Rob Clewley modified a comment on discussion Help

    HI All, I am trying to solve a simple demo problem, a Lotka-Volterra model defined...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Oh, but if you're using the latest git master, we do have a known bug involving a...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, since Dopri appears to be working, maybe you don't have a working gfortran or...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    That's not what it means: the precision is not related to the magnitude of the values...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    This is what you need: from PyDSTool import * V = Par(0.0001, 'V') Km = Par(1.0,...

  • Rob Clewley Rob Clewley modified a comment on discussion Help

    Hi All, A total newbie question. I have model defined through: V = Par(0.0001, 'V')...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Just on the code insert issue: yes, that args option is for python only. Hacking...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Alexandre, You're not comparing apples with apples, here. The PyCont example is...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    It's kind of hidden inside the complexity of the IF_delaynet example, so here's a...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    You can perform fixed point analysis and related continuation with PyCont for a discrete...

  • Rob Clewley Rob Clewley modified a comment on discussion Help

    1) & 2) Yes, makeModelInfoEntry is where that declaration happens, which is necessary...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    1 & 2. Yes, makeModelInfoEntry is where that declaration happens, which is necessary...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Yes, precision is part of the issue. The error actually reported for test 2 is at...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Yes, primarily because it's set for t=30 which is also the end of the integration...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi. You defined the end of your hybrid trajectory to be 'time' not your actual time-based...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    If you're saying that the data is externally produced and imported, then this can...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    It's an incorrect cross-reference of p in variable c: >>> print(channel_KCa.variables['c'].spec)...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, I'm sure we can resolve this but it's a bit cryptic without examples. Could you...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi Andrew, I hope Drew can answer this because I don't know how he chose to do it....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Yes, this is a known problem with Anaconda on Win 7 and 10: see this forum post....

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, yes there's a way to multi-D matrix-based ODEs if you first write it out as one...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    I'm confused. The files you are attached don't refer to the Mathieu equations you...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Please add the files as attachments (see the option beside your "Post" button) or...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Hi, I'm not a bifurcation expert but I know that LPC and GH bifurcations can be found...

  • Rob Clewley Rob Clewley posted a comment on discussion Help

    Although it is much slower than the other integrators, the Vode option at least lets...

1 >