Instructions to use stefan-it/umt5-small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use stefan-it/umt5-small with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("stefan-it/umt5-small") model = AutoModelForSeq2SeqLM.from_pretrained("stefan-it/umt5-small") - Notebooks
- Google Colab
- Kaggle
Unimax sampler
#1
by StephennFernandes - opened
hey @stefan-it i really appreciate all the hard work for merging umT5 to HF.
The umT5 model comes with a new sampling strategy called unimax that supposed to integrate with seqio on sampling dataset mixture.
was hoping if you have a way to implement the same or have link/resources for the same, as i believe the t5x and seqio repo haven't released the official code.