| language: | |
| - nl | |
| tags: | |
| - punctuation prediction | |
| - punctuation | |
| datasets: sonar | |
| license: mit | |
| widget: | |
| - text: "hervatting van de zitting ik verklaar de zitting van het europees parlement die op vrijdag 17 december werd onderbroken te zijn hervat" | |
| example_title: "Dutch Sample" | |
| metrics: | |
| - f1 | |
| This model predicts the punctuation of Dutch texts. We developed it to restore the punctuation of transcribed spoken language. | |
| This multilanguage model was trained on the [SoNaR Dataset](http://hdl.handle.net/10032/tm-a2-h5). | |
| The model restores the following punctuation markers: **"." "," "?" "-" ":"** | |
| ## Sample Code | |
| We provide a simple python package that allows you to process text of any length. | |
| ## Install | |
| To get started install the package from [pypi](https://pypi.org/project/deepmultilingualpunctuation/): | |
| ```bash | |
| pip install deepmultilingualpunctuation | |
| ``` | |
| ### Restore Punctuation | |
| ```python | |
| from deepmultilingualpunctuation import PunctuationModel | |
| model = PunctuationModel(model="oliverguhr/fullstop-dutch-sonar-punctuation-prediction") | |
| text = "hervatting van de zitting ik verklaar de zitting van het europees parlement die op vrijdag 17 december werd onderbroken te zijn hervat" | |
| result = model.restore_punctuation(text) | |
| print(result) | |
| ``` | |
| **output** | |
| > hervatting van de zitting. ik verklaar de zitting van het europees parlement, die op vrijdag 17 december werd onderbroken, te zijn hervat. | |
| ### Predict Labels | |
| ```python | |
| from deepmultilingualpunctuation import PunctuationModel | |
| model = PunctuationModel() | |
| text = "hervatting van de zitting ik verklaar de zitting van het europees parlement die op vrijdag 17 december werd onderbroken te zijn hervat" | |
| clean_text = model.preprocess(text) | |
| labled_words = model.predict(clean_text) | |
| print(labled_words) | |
| ``` | |
| **output** | |
| > [['hervatting', '0', 0.99998724], ['van', '0', 0.9999784], ['de', '0', 0.99991274], ['zitting', '.', 0.6771242], ['ik', '0', 0.9999466], ['verklaar', '0', 0.9998566], ['de', '0', 0.9999783], ['zitting', '0', 0.9999809], ['van', '0', 0.99996245], ['het', '0', 0.99997795], ['europees', '0', 0.9999783], ['parlement', ',', 0.9908242], ['die', '0', 0.999985], ['op', '0', 0.99998224], ['vrijdag', '0', 0.9999831], ['17', '0', 0.99997985], ['december', '0', 0.9999827], ['werd', '0', 0.999982], ['onderbroken', ',', 0.9951485], ['te', '0', 0.9999677], ['zijn', '0', 0.99997723], ['hervat', '.', 0.9957053]] | |
| ## Results | |
| The performance differs for the single punctuation markers as hyphens and colons, in many cases, are optional and can be substituted by either a comma or a full stop. The model achieves the following F1 scores: | |
| | Label | F1 Score | | |
| | ------------- | -------- | | |
| | 0 | 0.985816 | | |
| | . | 0.854380 | | |
| | ? | 0.684060 | | |
| | , | 0.719308 | | |
| | : | 0.696088 | | |
| | - | 0.722000 | | |
| | macro average | 0.776942 | | |
| | micro average | 0.963427 | | |
| ## References | |
| TBD | |