mqt5-small

For use with the pyterrier-quality package.

A version of mt5-small trained as a passage quality estimation model using the approach described in this paper, over the following datasets: msmarco-passage, mmarco/de, mmarco/es, mmarco/fr, mmarco/id, mmarco/it, mmarco/pt, mmarco/ru, mmarco/v2/ar, mmarco/v2/de, mmarco/v2/dt, mmarco/v2/es, mmarco/v2/fr, mmarco/v2/hi, mmarco/v2/id, mmarco/v2/it, mmarco/v2/ja, mmarco/v2/pt, mmarco/v2/ru, mmarco/v2/vi, mmarco/v2/zh, mmarco/zh, neumarco/fa, neumarco/ru, neumarco/zh

>>> from pyterrier_quality import QualT5
>>> qt5 = QualT5('pyterrier-quality/mqt5-small')
>>> qt5([
...   {'docno': '0', 'text': 'bla bla bla'},
...   {'docno': '0', 'text': 'The presence of communication amid scientific minds was equally important to the success of the Manhattan Project as scientific intellect was. The only cloud hanging over the impressive achievement of the atomic researchers and engineers is what their success truly meant; hundreds of thousands of innocent lives obliterated.'},
... ])
  docno                                               text   quality
0     0                                        bla bla bla -1.406250
1     0  The presence of communication amid scientific ... -0.828125
>>> # A larger quality score means higher quality