Instructions to use facebook/mbart-large-cc25 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/mbart-large-cc25 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="facebook/mbart-large-cc25")# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("facebook/mbart-large-cc25") model = AutoModelForSeq2SeqLM.from_pretrained("facebook/mbart-large-cc25") - Inference
- Notebooks
- Google Colab
- Kaggle
Add TF weights
#2
by joaogante - opened
Model converted by the transformers' pt_to_tf CLI. All converted model outputs and hidden layers were validated against its PyTorch counterpart.
Maximum crossload output difference=4.578e-05; Maximum crossload hidden layer difference=1.465e-03;
Maximum conversion output difference=4.578e-05; Maximum conversion hidden layer difference=1.465e-03;
CAUTION: The maximum admissible error was manually increased to 0.002!
joaogante changed pull request status to merged