Instructions to use facebook/m2m100_418M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/m2m100_418M with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("facebook/m2m100_418M") model = AutoModelForSeq2SeqLM.from_pretrained("facebook/m2m100_418M") - Notebooks
- Google Colab
- Kaggle
Batching with different language pairs
#17
by acraber - opened
I know this is a long-shot, but has anyone found a way to do batch processing with different language pairs? I like that the translator has so many languages, but if I want to translate both es-en and en-es, I have to do them on separate batches. This adds a lot of overhead if I'm serving requests, as I need to run all the inferences separately.
Even if it has to be set to one language for the input or one language for the target, it would still help a lot.