Mark Bates
Mark Bates
Have you checked the return value of the call to gpufit(), and the output_states parameter?
This looks like a bug. Thanks for the feedback.
For debugging purposes, it may be useful to implement the model within Cpufit, first. Then, you can use a standard debugger without having to run the code on the GPU.
> I think that having one additional output to LMFit::LMFit with the values of the fitted curve can be useful to avoid computing again the model from the estimated parameters....
We will look at this possibility. Having an alternate function call may be a viable solution.
Leaving this issue open as a reminder that extending the Gpufit interface to return function values could be an option in a future release.
Sounds great. The best way to do this is fork the repository, include your changes, and submit a pull request.
This is a reminder for a future improvement to the library.
This is becoming increasingly important.
Don't have time to run this example at the moment, but did you try adding noise to the data?