Context:

https://groups.google.com/forum/?utm_medium=email&utm_source=footer#!msg/ceres-solver/cPMgCxiMSig/ff9WlJI0CAAJ

Could you please reference the papers used for the example implementations? That could possibly clear up a lot of confusion regarding the energy specifications.

Thanks!

It seems that there is a bug with the call to gpufit within Matlab when including the user_info parameter. Using the included linear_1d model (which utilizes the user_info parameter), I created a simple program in Matlab to model the equation y=x from x=0 to x=10 and called gpufit on the data. This should return the parameters 0 and 1, but results in 4.3467 and 0.8711 instead.

Additionally, if

The test suite contains a set of points to fit without indicating where those numbers came from. Plotting them in Excel suggests that the 3rd value (y = 21.119) may be a typo (instead of y = 2.119).

The value added by a test suite is only as good as the tests are meaningful. Does anyone know where these data came from? Were they measured from some physical process? Were they generated by a