---
license: apache-2.0
datasets:
- crodri/ceil
language:
- ca
metrics:
- type: f1
  value: 0.836
- type: precision
  value: 0.82069
- type: recall
  value: 0.8523
pipeline_tag: token-classification

widget:

- text: "L'ANC vol que l'11 de setembre al Passeig de Gràcia sigui una fita enguany."

- text: "El Martí llegeix el Cavall Fort."

- text: "El raper nord-americà Travis Scott ha gravat el videoclip de la seva cançó 'Circus Maximus' amb els Castellers de Vilafranca. Segons ha publicat la 'Revista Castells' i ha confirmat l'Agència Catalana de Notícies (ACN), el rodatge es va fer el 2 de juliol a la Tarraco Arena Plaça (TAP) de Tarragona." 


---

# Catalan BERTa (RoBERTa-large) finetuned for Named Entity Recognition.

## Table of Contents
<details>
<summary>Click to expand</summary>

- [Model description](#model-description)
- [Intended uses and limitations](#intended-uses-and-limitations)
- [How to Use](#how-to-use)
- [Training](#training)
  - [Training data](#training-data)
  - [Training procedure](#training-procedure)
- [Evaluation](#evaluation)
   - [Variable and metrics](#variable-and-metrics)
   - [Evaluation results](#evaluation-results)
- [Additional information](#addional-information)
  - [Author](#author)
  - [Contact information](#contact-information)
  - [Copyright](#copyright)
  - [Licensing information](#licensing-information)
  - [Funding](#funding)
  - [Citing information](#citing-information)
  - [Disclaimer](#disclaimer))
</details>


## Model description

The **multiner** is a Named Entity Recognition (NER) model for the Catalan language fine-tuned from the [BERTa] model, a [RoBERTa](https://arxiv.org/abs/1907.11692) base model pre-trained on a medium-size corpus collected from publicly available corpora and crawlers (check the BERTa model card for more details).

It has been trained with a dataset that contains 9 main types and 52 subtypes on all kinds of short texts, with almost 59K documents. 

## Intended uses and limitations


## How to use


## Limitations and bias
At the time of submission, no measures have been taken to estimate the bias embedded in the model. However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.

## Training
We used the NERC dataset in Catalan called [Catalan Entity Identification and Linking](https://huggingface.co/datasets/crodri/ceil) for training and evaluation.

## Evaluation 

Accuracy was calculated using the development set, and reflects the non-balanced nature of the dataset.

### Major types

| Type 	| Accuracy 	| num. Instances in dev set 	|
| ------ | ------ | ------ |
| CW | 0.842 | 4551 |
| GPE | 0.914 | 19751 |
| Other | 0.69 | 2824 |
| building | 0.736 | 2188 |
| event | 0.739 | 3000 |
| location | 0.819 | 3408 |
| organization | 0.895 | 17285 |
| person | 0.903 | 21689 |
| product | 0.64 | 1038 |


### Subtypes

| Type 	| Accuracy 	| num. Instances in dev set	|
| ------ | ------ | ------ |
| CW-broadcastprogram | 0.854 | 765 |
| CW-film | 0.809 | 549 |
| CW-music | 0.862 | 1027 |
| CW-other | 0.495 | 555 |
| CW-painting | 0.654 | 205 |
| CW-writtenart | 0.814 | 1450 |
| GPE | 0.914 | 19751 |
| Other | 0.69 | 2824 |
| building-airport | 0.733 | 176 |
| building-governmentfacility | 0.514 | 72 |
| building-hospital | 0.805 | 113 |
| building-hotel | 0.688 | 32 |
| building-other | 0.726 | 1585 |
| building-religious | 0.0 | 1 |
| building-restaurant | 0.458 | 48 |
| building-shops | 0.206 | 34 |
| building-sportsfacility | 0.74 | 127 |
| event-attack/terrorism/militaryconflict | 0.866 | 411 |
| event-disaster | 0.261 | 23 |
| event-other | 0.695 | 1069 |
| event-political | 0.527 | 444 |
| event-protest | 0.207 | 29 |
| event-sportsevent | 0.822 | 1024 |
| location-bodiesofwater | 0.865 | 673 |
| location-island | 0.457 | 140 |
| location-mountain | 0.781 | 515 |
| location-other | 0.757 | 1602 |
| location-park | 0.581 | 93 |
| location-road/railway/highway/transit | 0.805 | 385 |
| organization-education | 0.868 | 2097 |
| organization-government | 0.905 | 2939 |
| organization-media | 0.888 | 1963 |
| organization-onlinebusiness | 0.538 | 197 |
| organization-other | 0.788 | 4733 |
| organization-politicalparty | 0.956 | 2272 |
| organization-privatecompany | 0.849 | 1809 |
| organization-religious | 0.638 | 210 |
| organization-sportsteam | 0.946 | 1065 |
| person-actor/director | 0.797 | 1480 |
| person-artist/author | 0.853 | 5812 |
| person-athlete | 0.871 | 1306 |
| person-group | 0.485 | 699 |
| person-influencer | 0.0 | 17 |
| person-other | 0.811 | 8444 |
| person-politician | 0.863 | 3259 |
| person-scholar/scientist | 0.728 | 672 |
| product-E-device | 0.51 | 102 |
| product-clothing | 0.222 | 27 |
| product-consumer_good | 0.0 | 20 |
| product-food | 0.673 | 324 |
| product-other | 0.0 | 69 |
| product-software | 0.67 | 382 |
| product-vehicle | 0.825 | 114 |


## Additional information

### Author
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center (langtech@bsc.es)

### Contact information
For further information, send an email to langtech@bsc.es

### Copyright
Copyright (c) 2021 Text Mining Unit at Barcelona Supercomputing Center 

### Licensing Information
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)

### Funding
This work was funded by the [Departament de la Vicepresidència i de Polítiques Digitals i Territori de la Generalitat de Catalunya](https://politiquesdigitals.gencat.cat/ca/inici/index.html#googtrans(ca|en) within the framework of [Projecte AINA](https://politiquesdigitals.gencat.cat/ca/economia/catalonia-ai/aina).

### Citation information



### Disclaimer

<details>
<summary>Click to expand</summary>

The models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions.

When third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their use and, in any event, to comply with applicable regulations, including regulations regarding the use of Artificial Intelligence.

In no event shall the owner and creator of the models (BSC – Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties of these models.