-
**Bug Report Checklist**
- [x] I provided code that demonstrates a minimal reproducible example.
- [x] I confirmed bug exists on the latest mainline of AutoGluon via source install.
- [ ] I c…
-
### 🐛 Describe the bug
frexp operator with out variant causes frame tracing error when run with dynamic=None.
If we disable dynamic shape i.e. dynamic=False for torch.compile then it works fine
use…
-
### 🐛 Describe the bug
Hi, when I use `autocast` I get numerical differences in the output of basic pytorch operators (in the attached example, `nn.Linear`) depending on whether or not `vmap` is used…
-
## 🐛 Bug
## To Reproduce
I run the [tutorial code](https://docs.dgl.ai/en/latest/guide/training-graph.html), but errors occur.
```python
import dgl.data
import torch.nn.functional as …
-
### 🐛 Describe the bug
When I run
```
./benchmarks/dynamo/torchbench.py --training --performance --no-skip --inductor --only moco
```
I get
```
File "/data/users/ezyang/a/pytorch/torch/…
-
![螢幕擷取畫面 2023-11-24 102454](https://github.com/sdbds/SVD-webui/assets/151893693/cd331cd0-6e5e-4ebd-9e78-36b54c4014d8)
How should this problem be solved?
-
Hi,
I created an environment that should contain everything needed to run your scripts:
conda create -n igmodel python=3.9
conda activate igmodel
conda install -c conda-forge cudatoolkit=11.8 …
CLG68 updated
4 months ago
-
### 🚀 The feature
A dataloader which communicates with its workers via torch.distributed.rpc API.
### Motivation, pitch
Presently, process-based workers for Dataloader mean the workers live o…
-
### 🐛 Describe the bug
Google's Gemma cannot be exported using `torch.export.export` due to a user defined type.
Fake Tensor is used because the model is very big.
A context manager is also used …
-
### 🐛 Describe the bug
mixnet_l amp multiple threads static shape default wrapper performance regression
name
batch_size_new
speed_up_new
inductor_new
e…