COMET

COMET copied to clipboard

Hi, I am trying to do some ablation studies on my custom pSQM dataset. I am constrained for compute and memory and cannot train the large (L/XL) or explainable models...

aenaliph

For the dataset shared here: https://huggingface.co/datasets/RicardoRei/wmt-mqm-human-evaluation the data summary says the score is: MQM score. Sample row below: > 1776 en-de He said: "I know of several other guys over...

aenaliph

Cannot use `load_from_checkpoint` in an offline environment

8

### Motivation Due to the policy of my company, I cannot access to an online server (e.g., huggingface). So I tried to use `load_from_checkpoint` function after I manually downloaded `wmt22-comet-da`...

zzaebok

This PR adds `local_files_only` parameter to `load_from_checkpoint` method as discussed in https://github.com/Unbabel/COMET/issues/225 The `local_files_only` parameter is created in the constructor of each encoder and CometModel. I found that even after...

zzaebok

[QUESTION] how to enable multi-gpu when calling the predict method

1

The predict method is called by setting gpus=2, but only one gpu is used. `model.predict(samples=samples,batch_size=2, gpus=2, devices=2, accelerator="gpu", strategy="ddp")`

cumtsunzhenxing

Bumps [protobuf](https://github.com/protocolbuffers/protobuf) from 4.25.5 to 4.25.8. Commits a4cbdd3 Updating version.json and repo version numbers to: 25.8 29445be Merge pull request #21880 from shaod2/py-25 cc13b69 Remove debugging code and add EOLs...

dependabot[bot]

Bumps [requests](https://github.com/psf/requests) from 2.32.3 to 2.32.4. Release notes Sourced from requests's releases. v2.32.4 2.32.4 (2025-06-10) Security CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve...

dependabot[bot]

jsonargparse==3.13.1 required for comet is incompatible with Python>=3.12

2

## 🐛 Bug ### To Reproduce Hello ! I attempt to use the `--enable-context` feature, so I tried to install the recent version of comet via `pip install unbabel-comet` in...

ziqianPeng

## 🐛 Bug Using torch.set_default_device to set anything else than "cpu" leads to a ValueError when calling model.predict. ### To Reproduce I'm using the code sample provided on the huggingface...

r-carpentier

Support for EuroLLM encoder

4

## 🚀 Feature Support for more recent encoders such as the one use for EuroLLM. ### Motivation First, thanks a lot for the great work on this framework! We have...

dmar1n