-
### Description
We had some issues with the FastAI dependency #2182. And realised that the fastai notebook is just a fancy wrapper on [PyTorch cartesian product](https://pytorch.org/docs/stable/gen…
-
### 🚀 The feature, motivation and pitch
**We are kicking off another community sprint!**
This community sprint resolves around extending the latest **GNNLLM features** from PyG 2.6.
This sprint…
-
As the previous discussion https://github.com/kubeflow/training-operator/pull/2024#discussion_r1521943323, I would suggest to introduce `torch.compile` in PyTorch examples, following the migration of …
-
### Problem
The sign "=" is not supported in Windows environment variables.
Thus, PYTORCH_CUDA_ALLOC_CONF=expandable_segments cannot be used on that platform.
### Solution
Could you please either …
-
Platforms: linux
This test was disabled because it is failing on main branch ([recent examples](https://torch-ci.com/failure?failureCaptures=%5B%22inductor%2Ftest_cpu_cpp_wrapper.py%3A%3ADynamicSha…
-
## Describe the bug
After following the installation instructions (plus replacing phonemizer with https://github.com/justinjohn0306/phonemizer to make it work on Win 11), and using the same examples …
-
Input MSAs were truncated to be a single entry (duplicates of the input sequences) because leaving `msa:` blank causes errors for some reason.
```
>101
MGDWSALGRLLDKVQAYSTAGGKVWLSVLFIFRILLLGTAVESA…
-
### System Info
```Shell
- `Accelerate` version: 1.1.0
- Platform: Linux-6.8.0-45-generic-x86_64-with-glibc2.35
- `accelerate` bash location: .venv/bin/accelerate
- Python version: 3.11.10
- N…
-
**Describe the bug**
Megatron-LM doesn't compatible with transformer-engine 1.13.
in transformer-engine:
https://github.com/NVIDIA/TransformerEngine/blob/2643ba1df43397cc84c9da5fe719a66d87ad9a0a/tr…
-
I-JEPA is compatible with timm since https://github.com/lightly-ai/lightly/pull/1612
We should update the I-JEPA examples to use timm instead of torchvision:
- [ ] Update `examples/pytorch/ijepa…