Fine tuning Model
Hello!
I'm attempting to perform fine-tuning on this English-to-Spanish translation model. I have sentence pairs stored in xlsx files for both English and Spanish. However, the process is not as straightforward as with the OPUS-mt-en-es model, where I could load the model using the following code:
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
model_name = "Helsinki-NLP/opus-mt-en-es"
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
Does anyone have a tutorial, preferably in Google Colab, on how to perform fine-tuning on the Seamless-M4T model?
Did you resolve fine-tuning?
If you find any answer to that, please share it with us here.
Hi guys, the source repo contains some code for finetuning: https://github.com/facebookresearch/seamless_communication/blob/main/src/seamless_communication/cli/m4t/finetune/README.md