-
related #7891
We could add a helper function that computes first and second derivatives for only one element of a partition of the parameter space.
We have several applications where we can comput…
-
We currently don't have any numdiff fallback if some analytical derivatives are missing in the model
I'm trying out MLEInlfuence(res) where res is a NegativeBinomialP results instance.
but I get
…
-
- [ ] Write code to compute covariance for generic XY and WZ tracers and test on e.g. TT assigned to @reneehlozek
- [ ] write code to compute numerical derivatives for the Cl terms assigned to @rene…
-
This issue is extracted from the discussion in #6998.
Currently we have the concepts of "configuration" and "velocity" variables mixed up with "continuous" and "discrete". The generalized configura…
-
`model.d_phase_d_param()` returns a derivative whose mean is zero. This is wrong if one is working with absolute phases - if the model contains `TZRMJD`, all derivatives should be exactly zero there. …
-
Good idea for checking analytical derivatives to high accuracy.
* also get an estimation of the numerical error
-
Hello,
First of all, many thanks for this library - it is great! I am trying to implement one idea that I had and I am constantly finding that I have no issues on forward simulation using `odeint_…
-
Thoughts after questions from AmpLab staff Jey Kottalam:
So far we have been providing analytical expressions for the 4th spatial derivatives for the covariance kernel, it is possible for us to explor…
-
I am interested in a DG solver for hamilton-jacobi equations (of the form $`\varphi_t + H(\nabla \varphi) = 0`$) as they arise in level-set methods (e.g. $`H(\nabla \varphi) = a \| \nabla \varphi \|`$…
-
tl;dr: While JAX clamps indices to the array bounds when evaluating array indexing operations, the same logic does not appear to be applied when back-propagating derivatives through the indexing opera…