search

tBuLi / symfit

Symbolic Fitting; fitting as it should be.
http://symfit.readthedocs.org
MIT License
235 stars 19 forks source link

Fixed #161. #163

Closed tBuLi closed 6 years ago

tBuLi commented 6 years ago

In case of failure to estimate the shape of the jacobian, we check if this is becuase the shape of the data-dimension is not properly estimated and adjust if necessary.

tBuLi commented 6 years ago

Yes, they failed without the fix. I agree that these tests are not great, but I struggle to come up with a better way to test for this.

I thought about not changing the inheritance, but since there is actually a lot of self.dependent_data-like statements in HasCovarianceMatrix, it already assumed that it has data. So I think this way also takes care of the implicit dependency there.

Taking the shape from W itself is not a bad idea, but I went this way because of the argument above. Do you agree with the rationale?

pckroon commented 6 years ago

Yes, makes sense :) Nice work