Update README.md
Browse files
README.md
CHANGED
@@ -1,248 +1,131 @@
|
|
1 |
---
|
2 |
-
|
3 |
-
language:
|
4 |
-
|
5 |
-
- ca
|
6 |
-
|
7 |
license: apache-2.0
|
8 |
-
|
9 |
-
tags:
|
10 |
-
|
11 |
-
- "catalan"
|
12 |
-
|
13 |
-
- "named entity recognition"
|
14 |
-
|
15 |
-
- "ner"
|
16 |
-
|
17 |
-
- "CaText"
|
18 |
-
|
19 |
-
- "Catalan Textual Corpus"
|
20 |
-
|
21 |
datasets:
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
metrics:
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
-
|
31 |
-
|
32 |
-
|
33 |
-
type: token-classification
|
34 |
-
dataset:
|
35 |
-
type: crodri/ceil
|
36 |
-
name: ancora-ca-ner
|
37 |
-
metrics:
|
38 |
-
- type: f1
|
39 |
-
value: 0.836
|
40 |
-
- type: precision
|
41 |
-
value: 0.82069
|
42 |
-
- type: recall
|
43 |
-
value: 0.8523
|
44 |
|
45 |
widget:
|
46 |
|
47 |
-
- text: "El raper nord-americà Travis Scott ha gravat el videoclip de la seva cançó
|
48 |
|
49 |
- text: "L'ANC vol que l'11 de setembre al Passeig de Gracia sigui una fita enguany."
|
50 |
|
51 |
- text: "El Martí llegeix el Cavall Fort."
|
52 |
|
|
|
53 |
|
54 |
-
#
|
55 |
-
|
56 |
-
<!-- Provide a quick summary of what the model is/does. -->
|
57 |
-
|
58 |
-
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
|
59 |
-
|
60 |
-
## Model Details
|
61 |
-
|
62 |
-
### Model Description
|
63 |
-
|
64 |
-
<!-- Provide a longer summary of what this model is. -->
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
- **Developed by:** [More Information Needed]
|
69 |
-
- **Shared by [optional]:** [More Information Needed]
|
70 |
-
- **Model type:** [More Information Needed]
|
71 |
-
- **Language(s) (NLP):** [More Information Needed]
|
72 |
-
- **License:** [More Information Needed]
|
73 |
-
- **Finetuned from model [optional]:** [More Information Needed]
|
74 |
-
|
75 |
-
### Model Sources [optional]
|
76 |
-
|
77 |
-
<!-- Provide the basic links for the model. -->
|
78 |
-
|
79 |
-
- **Repository:** [More Information Needed]
|
80 |
-
- **Paper [optional]:** [More Information Needed]
|
81 |
-
- **Demo [optional]:** [More Information Needed]
|
82 |
-
|
83 |
-
## Uses
|
84 |
-
|
85 |
-
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
86 |
-
|
87 |
-
### Direct Use
|
88 |
-
|
89 |
-
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
90 |
-
|
91 |
-
[More Information Needed]
|
92 |
-
|
93 |
-
### Downstream Use [optional]
|
94 |
-
|
95 |
-
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
96 |
-
|
97 |
-
[More Information Needed]
|
98 |
-
|
99 |
-
### Out-of-Scope Use
|
100 |
-
|
101 |
-
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
102 |
-
|
103 |
-
[More Information Needed]
|
104 |
-
|
105 |
-
## Bias, Risks, and Limitations
|
106 |
-
|
107 |
-
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
108 |
-
|
109 |
-
[More Information Needed]
|
110 |
-
|
111 |
-
### Recommendations
|
112 |
-
|
113 |
-
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
114 |
-
|
115 |
-
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
116 |
-
|
117 |
-
## How to Get Started with the Model
|
118 |
-
|
119 |
-
Use the code below to get started with the model.
|
120 |
-
|
121 |
-
[More Information Needed]
|
122 |
-
|
123 |
-
## Training Details
|
124 |
-
|
125 |
-
### Training Data
|
126 |
-
|
127 |
-
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
128 |
-
|
129 |
-
[More Information Needed]
|
130 |
-
|
131 |
-
### Training Procedure
|
132 |
-
|
133 |
-
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
134 |
-
|
135 |
-
#### Preprocessing [optional]
|
136 |
-
|
137 |
-
[More Information Needed]
|
138 |
-
|
139 |
-
|
140 |
-
#### Training Hyperparameters
|
141 |
-
|
142 |
-
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
143 |
-
|
144 |
-
#### Speeds, Sizes, Times [optional]
|
145 |
-
|
146 |
-
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
147 |
-
|
148 |
-
[More Information Needed]
|
149 |
-
|
150 |
-
## Evaluation
|
151 |
-
|
152 |
-
<!-- This section describes the evaluation protocols and provides the results. -->
|
153 |
-
|
154 |
-
### Testing Data, Factors & Metrics
|
155 |
-
|
156 |
-
#### Testing Data
|
157 |
-
|
158 |
-
<!-- This should link to a Data Card if possible. -->
|
159 |
-
|
160 |
-
[More Information Needed]
|
161 |
-
|
162 |
-
#### Factors
|
163 |
-
|
164 |
-
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
165 |
-
|
166 |
-
[More Information Needed]
|
167 |
-
|
168 |
-
#### Metrics
|
169 |
-
|
170 |
-
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
171 |
-
|
172 |
-
[More Information Needed]
|
173 |
-
|
174 |
-
### Results
|
175 |
-
|
176 |
-
[More Information Needed]
|
177 |
-
|
178 |
-
#### Summary
|
179 |
-
|
180 |
-
|
181 |
-
|
182 |
-
## Model Examination [optional]
|
183 |
-
|
184 |
-
<!-- Relevant interpretability work for the model goes here -->
|
185 |
-
|
186 |
-
[More Information Needed]
|
187 |
-
|
188 |
-
## Environmental Impact
|
189 |
-
|
190 |
-
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
191 |
-
|
192 |
-
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
193 |
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
- **Compute Region:** [More Information Needed]
|
198 |
-
- **Carbon Emitted:** [More Information Needed]
|
199 |
|
200 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
201 |
|
202 |
-
### Model Architecture and Objective
|
203 |
|
204 |
-
|
205 |
|
206 |
-
|
207 |
|
208 |
-
|
209 |
|
210 |
-
#### Hardware
|
211 |
|
212 |
-
|
213 |
|
214 |
-
#### Software
|
215 |
|
216 |
-
|
|
|
217 |
|
218 |
-
##
|
|
|
219 |
|
220 |
-
|
|
|
221 |
|
222 |
-
|
|
|
|
|
|
|
|
|
|
|
223 |
|
224 |
-
[
|
225 |
|
226 |
-
|
227 |
|
228 |
-
|
|
|
229 |
|
230 |
-
|
|
|
231 |
|
232 |
-
|
|
|
233 |
|
234 |
-
|
|
|
235 |
|
236 |
-
|
|
|
237 |
|
238 |
-
|
239 |
|
240 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
241 |
|
242 |
-
|
243 |
|
244 |
-
|
|
|
245 |
|
246 |
-
|
247 |
|
|
|
248 |
|
|
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
datasets:
|
4 |
+
- crodri/ceil
|
5 |
+
language:
|
6 |
+
- ca
|
7 |
metrics:
|
8 |
+
- type: f1
|
9 |
+
value: 0.836
|
10 |
+
- type: precision
|
11 |
+
value: 0.82069
|
12 |
+
- type: recall
|
13 |
+
value: 0.8523
|
14 |
+
pipeline_tag: token-classification
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
widget:
|
17 |
|
18 |
+
- 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."
|
19 |
|
20 |
- text: "L'ANC vol que l'11 de setembre al Passeig de Gracia sigui una fita enguany."
|
21 |
|
22 |
- text: "El Martí llegeix el Cavall Fort."
|
23 |
|
24 |
+
---
|
25 |
|
26 |
+
# Catalan BERTa (RoBERTa-large) finetuned for Named Entity Recognition.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
27 |
|
28 |
+
## Table of Contents
|
29 |
+
<details>
|
30 |
+
<summary>Click to expand</summary>
|
|
|
|
|
31 |
|
32 |
+
- [Model description](#model-description)
|
33 |
+
- [Intended uses and limitations](#intended-uses-and-limitations)
|
34 |
+
- [How to Use](#how-to-use)
|
35 |
+
- [Training](#training)
|
36 |
+
- [Training data](#training-data)
|
37 |
+
- [Training procedure](#training-procedure)
|
38 |
+
- [Evaluation](#evaluation)
|
39 |
+
- [Variable and metrics](#variable-and-metrics)
|
40 |
+
- [Evaluation results](#evaluation-results)
|
41 |
+
- [Additional information](#addional-information)
|
42 |
+
- [Author](#author)
|
43 |
+
- [Contact information](#contact-information)
|
44 |
+
- [Copyright](#copyright)
|
45 |
+
- [Licensing information](#licensing-information)
|
46 |
+
- [Funding](#funding)
|
47 |
+
- [Citing information](#citing-information)
|
48 |
+
- [Disclaimer](#disclaimer))
|
49 |
+
</details>
|
50 |
|
|
|
51 |
|
52 |
+
## Model description
|
53 |
|
54 |
+
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).
|
55 |
|
56 |
+
## Intended uses and limitations
|
57 |
|
|
|
58 |
|
59 |
+
## How to use
|
60 |
|
|
|
61 |
|
62 |
+
## Limitations and bias
|
63 |
+
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.
|
64 |
|
65 |
+
## Training
|
66 |
+
We used the NER dataset in Catalan called [Ancora-ca-ner](https://huggingface.co/datasets/projecte-aina/ancora-ca-ner) for training and evaluation.
|
67 |
|
68 |
+
## Evaluation
|
69 |
+
We evaluated the _roberta-base-ca-cased-ner_ on the Ancora-ca-ner test set against standard multilingual and monolingual baselines:
|
70 |
|
71 |
+
| Model | Ancora-ca-ner (F1)|
|
72 |
+
| ------------|:-------------|
|
73 |
+
| roberta-base-ca-cased-ner | **88.13** |
|
74 |
+
| mBERT | 86.38 |
|
75 |
+
| XLM-RoBERTa | 87.66 |
|
76 |
+
| WikiBERT-ca | 77.66 |
|
77 |
|
78 |
+
For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club).
|
79 |
|
80 |
+
## Additional information
|
81 |
|
82 |
+
### Author
|
83 |
+
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center (bsc-temu@bsc.es)
|
84 |
|
85 |
+
### Contact information
|
86 |
+
For further information, send an email to aina@bsc.es
|
87 |
|
88 |
+
### Copyright
|
89 |
+
Copyright (c) 2021 Text Mining Unit at Barcelona Supercomputing Center
|
90 |
|
91 |
+
### Licensing Information
|
92 |
+
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
|
93 |
|
94 |
+
### Funding
|
95 |
+
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).
|
96 |
|
97 |
+
### Citation information
|
98 |
|
99 |
+
If you use any of these resources (datasets or models) in your work, please cite our latest paper:
|
100 |
+
```bibtex
|
101 |
+
@inproceedings{armengol-estape-etal-2021-multilingual,
|
102 |
+
title = "Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? {A} Comprehensive Assessment for {C}atalan",
|
103 |
+
author = "Armengol-Estap{\'e}, Jordi and
|
104 |
+
Carrino, Casimiro Pio and
|
105 |
+
Rodriguez-Penagos, Carlos and
|
106 |
+
de Gibert Bonet, Ona and
|
107 |
+
Armentano-Oller, Carme and
|
108 |
+
Gonzalez-Agirre, Aitor and
|
109 |
+
Melero, Maite and
|
110 |
+
Villegas, Marta",
|
111 |
+
booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
|
112 |
+
month = aug,
|
113 |
+
year = "2021",
|
114 |
+
address = "Online",
|
115 |
+
publisher = "Association for Computational Linguistics",
|
116 |
+
url = "https://aclanthology.org/2021.findings-acl.437",
|
117 |
+
doi = "10.18653/v1/2021.findings-acl.437",
|
118 |
+
pages = "4933--4946",
|
119 |
+
}
|
120 |
+
```
|
121 |
|
122 |
+
### Disclaimer
|
123 |
|
124 |
+
<details>
|
125 |
+
<summary>Click to expand</summary>
|
126 |
|
127 |
+
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.
|
128 |
|
129 |
+
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.
|
130 |
|
131 |
+
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.
|