ZincCat
ZincCat
Thanks for this great project!
This pull request adds the server_streaming.py example for the high-level API. The server_streaming.py file is created under the examples/high_level_api/ directory. It contains an implementation of a server streaming example using...
When working on a 3D problem, the memory usage of Vector_Test is too high compared to the actual sampled points. An example is shown below which caused crash. ```python test_fn...
The implementation of get_num_losses in `domain.py` is summing up the number of output names in constraints. Yet this is not the case when using custom loss function, e.g. when using...
The implementation is in here: https://github.com/zinccat/dinov2-patch
changed the normalization constant to list[float], removed squeeze in loss calculation step so a full batch can be trained without error
### Reminder - [x] I have read the above rules and searched the existing issues. ### Description Any plan to integrate verl to support training reasoning models? ### Pull Request...
When running `examples/hard_constraints.py` using two gpus, has following error: ```bash hfppl/hfppl/llms.py", line 244, in past_padded return torch.cat( ^^^^^^^^^^ RuntimeError: Expected all tensors to be on the same device, but found...
As mps is not currently supported for CTranslate2, is it possible to add a whisper.cpp based backend for better performance on apple metal devices? Thanks