Instructions to use Unbabel/XCOMET-XL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Unbabel/XCOMET-XL with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("translation", model="Unbabel/XCOMET-XL")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Unbabel/XCOMET-XL", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Support of numpy 2.x and protobuf 5.2x ?
#6
by raptorkwok - opened
Hello there,
When I tried to install unlabel-comet 2.2.6 (using pip install "unlabel-comet>=2.2.0"), I noticed it requires numpy 1.x and protobuf 4.x.
I had numpy 2.0.2 and protobuf 5.29.3 installed and had to downgrade in order to use this unlabel-comet 2.2.6