-
It looks like the L-BFGS-B solver doesn't work properly. I added a simple [test case](https://github.com/hugandr/CppNumericalSolvers/commit/f9f720749d862edb22e4868eb56123f6b1dbf6d1) in my fork:
The …
-
Update from Dimitri, Oct 2012: after talking with many developers and users, we should definitely make the inversion tools in SPECFEM3D more standard and more user friendly. Even if there are many opt…
-
we currently default to the reparameterization gradient if the `Variational` class implements `reparam`
however, if the `Inference` class does not support reparameterization gradients (e.g. `KLpq`) t…
-
Sorry if this turns up two times, I could not find my submission after the last time and I also forgot my name and email address.
My name is Gustaf Olsson (gustaf.olsson@lnu.se)
I am running Avogad…
-
Sorry if this turns up two times, I could not find my submission after the last time and I also forgot my name and email address.
My name is Gustaf Olsson (gustaf.olsson@lnu.se)
I am running Avogad…
cryos updated
8 years ago
-
I'm not sure where the best place to discuss Optim is so I'll post this here.
In my field, computational chemistry, it is common to enforce a maximum step size constraint at each optimization step, w…
-
When i ran example/trpo_gym.py, i got the results. Then i ran "scripts/submit_gym.py", but i got following error:
**raise error.Error("[%s] You didn't have any recorded training data in {}. Once you'v…
-
We have a couple of doc warnings currently. Many are of the type
```
/data/jenkins/workspace/odl-master-doc/odl/space/fspace.py:docstring of odl.space.fspace.FunctionSpace.__eq__:10: WARNING: more th…
-
Hi,
I am trying to run a regression task using the darch package but it looks the darch class only handles classification tasks. However, I do notice there is function called minimizeAutoencoder, wh…
-
`newton_cg` requires the function value to compute the step size (line search) but not to compute the descent direction: on [line 105](https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/…