-
Two consecutive runs of moldft can have very different convergence behaviour - it starts out the same but by the second or third iteration the energies are very different. They end up at very similar…
-
**Describe the bug**
On a DWI dataset, the JSON sidecar `FrameDuration` have only 1 element, but is written as an array, and the line is finished by a comma :
```
105,
…
-
Blocker: efficient nuclear norm calculation for large sparse matrices. See https://twitter.com/alexpghayes/status/1308439934986891266 for some useful references -- in particular it seems like I want a…
-
Example: https://mila.quebec/en/publications/
It would be nice to reuse the same code as in the Mila website. Not sure if that's 'easily' possible via RTD
-
# GMA WMT15 German-to-English with Transformer-Base : delta=2.5
@Vily1998
I've followed the guidance and trained the model for about 100 epoch using following set:
```
delta=2.5
python train.p…
-
### 🐛 Describe the bug
See https://hud.pytorch.org/ci/pytorch/pytorch/master?name_filter=xla
Starting after https://github.com/pytorch/pytorch/commit/84f168539725292d0b824c30d1d872fb30811803 CI runs…
-
https://arxiv.org/pdf/1710.09026v1.pdf
See lemma 1. If we want to optimize something like:
` |X - W|_F + lambda * |W|_T, `
Where the F is the Frobenius norm (taken only over known values), an…
-
Hi All,
I was running an older version of PyTorch ( - built from source) with FuncTorch ( - built from source), and somehow I've broken the older version of functorch. When I import functorch I get…
-
In GitLab by @pc6783 on Jun 21, 2021, 17:13
Currently, the centrepost (centerstack) nuclear heating (pnuc_cp_tf) is computed but not added to TF nuclear heating (ptfnuc). TF & centrepost nuclear heat…
-
Hello,
I'd like to implement a nuclear norm minimization approach for matrix completion using your package. This requires a constraint that `|X_{ij} - S_{ij}| < eps` where `X` is the incomplete mat…