README.md

metadata

language:
  - it
license: apache-2.0
tags:
  - italian
  - sequence-to-sequence
  - style-transfer
  - formality-style-transfer
datasets:
  - yahoo/xformal_it
widget:
  - text: maronn qualcuno mi spieg' CHECCOSA SUCCEDE?!?!
  - text: wellaaaaaaa, ma fraté sei proprio troppo simpatiko, grazieeee!!
  - text: nn capisco xke tt i ragazzi lo fanno
  - text: IT5 è SUPERMEGA BRAVISSIMO a capire tt il vernacolo italiano!!!
metrics:
  - rouge
  - bertscore
model-index:
  - name: it5-large-informal-to-formal
    results:
      - task:
          type: formality-style-transfer
          name: Informal-to-formal Style Transfer
        dataset:
          type: xformal_it
          name: XFORMAL (Italian Subset)
        metrics:
          - type: rouge1
            value: 0.663
            name: Avg. Test Rouge1
          - type: rouge2
            value: 0.477
            name: Avg. Test Rouge2
          - type: rougeL
            value: 0.645
            name: Avg. Test RougeL
          - type: bertscore
            value: 0.714
            name: Avg. Test BERTScore
            args:
              - model_type: dbmdz/bert-base-italian-xxl-uncased
              - lang: it
              - num_layers: 10
              - rescale_with_baseline: true
              - baseline_path: bertscore_baseline_ita.tsv
co2_eq_emissions:
  emissions: 51g
  source: Google Cloud Platform Carbon Footprint
  training_type: fine-tuning
  geographical_location: Eemshaven, Netherlands, Europe
  hardware_used: 1 TPU v3-8 VM

IT5 Base for Informal-to-formal Style Transfer 🧐

This repository contains the checkpoint for the IT5 Large model fine-tuned on Informal-to-formal style transfer on the Italian subset of the XFORMAL dataset as part of the experiments of the paper IT5: Large-scale Text-to-text Pretraining for Italian Language Understanding and Generation by Gabriele Sarti and Malvina Nissim.

A comprehensive overview of other released materials is provided in the gsarti/it5 repository. Refer to the paper for additional details concerning the reported scores and the evaluation approach.

Using the model

Model checkpoints are available for usage in Tensorflow, Pytorch and JAX. They can be used directly with pipelines as:

from transformers import pipelines

i2f = pipeline("text2text-generation", model='it5/it5-large-informal-to-formal')
i2f("nn capisco xke tt i ragazzi lo fanno")
>>> [{"generated_text": "non comprendo perché tutti i ragazzi agiscono così"}]

or loaded using autoclasses:

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("it5/it5-large-informal-to-formal")
model = AutoModelForSeq2SeqLM.from_pretrained("it5/it5-large-informal-to-formal")

If you use this model in your research, please cite our work as:

@article{sarti-nissim-2022-it5,
    title={IT5: Large-scale Text-to-text Pretraining for Italian Language Understanding and Generation},
    author={Sarti, Gabriele and Nissim, Malvina},
    journal={ArXiv preprint TBD},
    url={TBD},
    year={2022}
}