-
Just updated to OpenMM 7.3 for ChimeraX/ISOLDE, and my built-in demo/test model now runs agonizingly slow on minimization - about 5 minutes where it used to take 1-2 seconds. Each 1,000-round minimiza…
-
I am still not able to pass options to fminbox:
```
optimize(f, g!, lower_x, upper_x, x0, Fminbox(LBFGS()), Optim.Options(iterations=2, show_trace=true))
```
pawbz updated
6 years ago
-
When I read the code of logistic regression, I found that angel seems to fetch total weight vector in worker, and also create one dense gradient vector with the same length.
LRLearner.scala: 72L
…
-
Hi all,
First, thanks a lot for the great framework! It's been a fun experience learning and enabling fitting mixed effects models at scale.
I had a question about setting up configurations corr…
-
https://github.com/hjmshi/PyTorch-LBFGS/blob/6677e1f22c54b78f1481b517fd9e823484935de6/functions/LBFGS.py#L424
Instead of lengthy if-else sections, maybe better to have the different line search opt…
-
Accidently I tried to register an image to itself using the multi_scale optimizer, but the code breaks. It should return an identity map. (It works when single scale optimizer is used.)
-
Right now, the [toy examples](https://nbviewer.jupyter.org/github/dnouri/skorch/blob/master/notebooks/Basic_Usage.ipynb) break if the `torch.optim.LBFGS` optimizer is used (likely also some others, e.…
-
It would be nice to have at least one multivariate maximizer, akin to our Brent based scalar routine in
https://github.com/QuantEcon/QuantEcon.py/blob/master/quantecon/optimize/scalar_maximization.…
jstac updated
5 years ago
-
We have Windows 32bit testing in appveyor, which will be removed again in #3023
some unit tests where disabled for Win32
in one of the last appveyor runs a new error showed up, which should be inve…
-
## Issue description
When I am trying to run an optimizer on a separate variable inside a pytorch function, pytorch throws an error, ```element 0 of tensors does not require grad and does not have …