-
**Mod & Pack version** ftb-chunks-forge-1605.3.1-build.41.jar
**Forge / Fabric version, if applicable** 36.2.2
**How to reproduce the issue**
1. Start a new world to generate configs
2. …
-
> m. Why `RankEXACT`? Probably has no effect, but if so, we should say so. rule.generic.clear
_Originally posted by @rptb1 in https://github.com/Ravenbrook/mps/pull/34#discussion_r1123320599_
…
-
While working on PR #3864, I wanted to write a simple test in which I would run jobs under the `testsdexec` job-exec module (launching jobs under systemd), kill the broker, and then make sure the jobs…
chu11 updated
3 years ago
-
Thank you for your PyAutoFEP
According to PyAutoFEP Tutorial, there appeared the following wrong when I came to run bash runall.sh on the protein file from FXR_12-FXR_74 file. in other case, runni…
-
Hi, I am interested in using radical.pilot and it seems to work well for my purposes when it comes to running multiple independent Molecular Dynamics (MD) simulations on one node, within one HPC job. …
-
**Environment:**
1. Framework: Tensorflow/PyTorch
2. Framework version: all
3. Horovod version: all
4. MPI version: all
5. CUDA version: all
6. NCCL version: all
7. Python version: 3.*
8. Spar…
-
Hi,
I recently set up gsync to back up my music collection and all is working for the most part.
I do however have one problem, in that I cannot update existing files. No matter what I do every tim…
-
flarum version `1.3.1`
flarum `composer.json`
```
{
"name": "flarum/flarum",
"description": "Delightfully simple forum software.",
"type": "project",
"keywords": [
…
-
Specifically it failed at dealing with DTensor `MaskPartial` placement of sharded embedding.
This only happens when we do whole model compile.
TransformerBlock-level compilation (default) + separa…
-
Some tests, e.g., `long long` fail when using mpich and CUDA-aware GPU.