-
### Describe the bug
After training an epoch, it gets stuck in the middle of the second epoch, There is no problem with training with a single card (with command : `CUDA_VISIBLE_DEVICES=3 python tra…
-
### What happened + What you expected to happen
mamba install neuralforecast
and pip install neuralforecast
, get same errore
Stacktrace
```
Exception: Could not deserialize ATN with versi…
TDL77 updated
9 months ago
-
### Describe the bug
I am trying to install the required libraries in a kaggle notebook:
- !pip install super-gradients==3.1.0
- !pip install imutils
- !pip install roboflow
- !pip install pytu…
-
### 🐛 Describe the bug
Hi team, we are debugging some cuda graph memory leak in production. We’d like to know if the behavior is expected. We found some layers that boil down to
1)`torch.nn.Linear(i…
-
**Describe the bug**
See title
**What is the current behavior?**
**If the current behavior is a bug, please provide the steps to reproduce.**
1. Create a `TabNetPretrainer` instance, e.g…
-
### 🐛 Describe the bug
On cpu, `nn.LayerNorm` outputs a tensor of all zeros for single column batch in the `torch` nightly version. The all zero output tensor affects parameter updates.
This issu…
-
The status page of RTMPose showed that RTM-tiny model is 0.3G flops.
Just want to make sure it is flops or macs?I am using torchinfo to count and it showed that RTMPose-tiny model is 0.3G Macs(mult-a…
-
First of all, thank you for your work.
I would like to ask if the results obtained by that code can get the results described in the paper and if they can beat MTFAA and FRCRN in terms of objective s…
-
### 🐛 Describe the bug
I tried a very simple test program where a client tries to connect to a server ~using DDP~. The code hangs at init_process_group. Strangely it only hangs when I use gloo back…
-
### 🐛 Describe the bug
As stated in the title, the following crashes when using the `mps` device:
```python
ln = nn.LayerNorm((768,), elementwise_affine=True).to("mps")
ln(torch.randn(1, 77, 768…