-
Hi Mario,
Thanks for the implementation! I wanted to ask about the inference performance of allegro-jax, as it seems to be quite different from what I'm seeing in the official Pytorch implementatio…
-
I tried to visualize a StableHLO model with model-explorer, but got this error:
```
Failed to layout graph: RangeError: Maximum call stack size exceeded
```
Could you take a look of my model IR, I…
-
### 🐛 Describe the bug
We are unable to call allow_in_graph inside of a function being torch.compile'd with fullgraph=True and the error message is confusing. We should either allow this or improve t…
-
### 🐛 Describe the bug
When running a segmentation model (UNet) from segmentation_models_pytorch using PyTorch's torch._dynamo, I encountered an internal error.
The line that triggers the issue i…
-
When using PolicyRecordingBot, a call to the bot is established and recording starts correctly, but halfway the call, we receive an event that the call is deleted, and recording stops. The event also …
-
TRE Admins and all workspace users should be able to retrieve which users are assigned to a workspace.
This can be done by adding an additional route to `workspaces_shared_router`, for example: `/a…
-
Hi author,
Right now I download the pretrain weight from [Multiview Contrast](https://zenodo.org/record/7593637/files/mc_gearnet_edge.pth). I want to load this weight, and then pass a pdb file to the…
-
In my work to integrate #59445 into vscode-go, I am relying on file-change notifications to keep the test tree up to date as files change. @hyangah pointed out that the ideal solution would be to use …
-
### Feature Description
**Current Problem:**
We need to parallelize the process of creating the index of a knowledge graph. Currently, the code sequentially processes each call to the language model…
-
### 🚀 The feature, motivation and pitch
`torch.utils.checkpoint.checkpoint` has a strong limitation: there must be no graph break, and if it can't be satisfied, it will fall back to eager with this d…