-
Could the authors confirm or state if there is a mistake in any of the following statements about the code used for VAE:
**Statement 1**
The output of the `DecoderSamplerLayer` is equal to `z_enc` i…
botev updated
5 years ago
-
### Describe the bug
i am trying out pytorch 2.4.1, which seems to require minimal change to existing code ( mostly convert all .cuda to .xpu ), when running nn.CrossEntropyLoss, i always get Requi…
-
### 🐛 Describe the bug
I'm trying to register a custom CPP op with torch.library instead of the "old" way of directly calling the dlopened handle from an autograd function. I followed the [custom o…
-
### Checklist
- [ ] The issue exists after disabling all extensions
- [X] The issue exists on a clean installation of webui
- [ ] The issue is caused by an extension, but I believe it is caused by a …
-
They seems to have similar function.
eeyrw updated
3 months ago
-
### 🐛 Describe the bug
By default, `torch.cdist` currently does not accept half-precision datatypes as input, throwing a `RuntimeError: "cdist" not implemented for 'Half'`.
However, if the inputs …
-
### Your current environment information
Collecting environment information...
PyTorch version: 2.1.1+cu118
Is debug build: False
CUDA used to build PyTorch: 11.8
ROCM used to build PyTorch: N/A
…
-
# Prerequisites
Please answer the following questions for yourself before submitting an issue.
- [X] I am running the latest code. Development is very rapid so there are no tagged versions as of…
-
### 🐛 Describe the bug
Under specific inputs, torch._add_batch_dim throws a false INTERNAL ASSERT FAILED, accompanied by the message: "please report a bug to PyTorch."
minimal example:
```
im…
x0w3n updated
3 weeks ago
-
### Describe the bug
Training a model with an LSTM module causes a segmentation fault after loss.backward() has been called a random number of times. The number of times the training loop can run …