-
### Description
When I'm using `cupyx.scipy.linalg.expm` on a 2x2 complex matrix, there is a small (~3e-6) absolute error wrt the correct value and wrt `scipy.linalg.expm`.
### To Reproduce
…
-
Bonjour à tous,
rappel : rendu ce soir minuit - pas de retard accepté.
Sinon, nous avons eu quelques questions pour venir lundi :
pour ceux qui viennent de Paris, avec l'expérience que nou…
-
One thing we found productive is TFM can be fairly easily ported to be variable precision. This in turn saves a lot of memory at a small (trivial) cost in performance.
I can't contribute code (rig…
-
It appears that the same kind of problem has been addressed multiple times for various dialects via internal `NO_SUPPORT_TIMESTAMP_PRECISION` flags:
- `AbstractDataType.NO_SUPPORT_TIMESTAMP_PRECISI…
-
With container `nvcr.io/nvidia/nemo:24.07` (Pyxis/Enroot), ran in Slurm interactive mode with 1 GPU if I execute the command
```bash
python3 /opt/NeMo/examples/asr/transcribe_speech_parallel.py \
...…
-
safdie
Medium
# Division precision loss in `createDistribution` function lead to incorrect distribution of rewards in `GammaRewarder.sol`
### Summary
The `createDistribution` function in the contr…
-
We should support temperature, humidity etc. sensors that communicate via Modbus. Examples:
- https://www.amazon.co.uk/Temperature-Humidity-Precision-Transmitter-Compatible/dp/B0836J2J2C. I have one …
-
### Describe the bug
Running train_controlnet_flux.py with multiple gpus results in a NCCL timeout error after N iterations of train_dataset.map(). This error can be partially solved by initializing …
-
Prevent possible precision errors when transforming vertices be multiple matrices.
Go through all of the shaders and change things like:
```
vec4 positionWC = czm_inverseView * positionEC;
vec4 posi…
-
Thanks for this nice repo.
**Describe the bug**
Finetuning in 2:4 sparsity w4a16 example fails with multiple GPUs
**Expected behavior**
The finetuning step expected to train successfully with …