mmarimon commited on
Commit
6ecb283
1 Parent(s): 688048a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -26
README.md CHANGED
@@ -52,29 +52,37 @@ widget:
52
  # Catalan BERTa (roberta-base-ca) finetuned for Text Classification.
53
 
54
  ## Table of Contents
55
- - [Model Description](#model-description)
56
- - [Intended Uses and Limitations](#intended-uses-and-limitations)
57
- - [How to Use](#how-to-use)
 
 
 
 
58
  - [Training](#training)
59
- - [Training Data](#training-data)
60
- - [Training Procedure](#training-procedure)
61
  - [Evaluation](#evaluation)
62
- - [Variable and Metrics](#variable-and-metrics)
63
- - [Evaluation Results](#evaluation-results)
64
- - [Licensing Information](#licensing-information)
65
- - [Citation Information](#citation-information)
66
- - [Funding](#funding)
67
- - [Contributions](#contributions)
68
- - [Disclaimer](#disclaimer)
 
 
 
 
69
 
70
  ## Model description
71
  The **roberta-base-ca-cased-tc** is a Text Classification (TC) model for the Catalan language fine-tuned from the roberta-base-ca 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.
72
 
73
- ## Intended Uses and Limitations
74
 
75
  **roberta-base-ca-cased-tc** model can be used to classify texts. The model is limited by its training dataset and may not generalize well for all use cases.
76
 
77
- ## How to Use
78
 
79
  Here is how to use this model:
80
 
@@ -89,17 +97,21 @@ tc_results = nlp(example)
89
  pprint(tc_results)
90
  ```
91
 
 
 
 
 
92
  ## Training
93
 
94
  ### Training data
95
  We used the TC dataset in Catalan called [TeCla](https://huggingface.co/datasets/projecte-aina/tecla) for training and evaluation.
96
 
97
- ### Training Procedure
98
  The model was trained with a batch size of 16 and a learning rate of 5e-5 for 5 epochs. We then selected the best checkpoint using the downstream task metric in the corresponding development set and then evaluated it on the test set.
99
 
100
  ## Evaluation
101
 
102
- ### Variable and Metrics
103
 
104
  This model was finetuned maximizing accuracy.
105
 
@@ -115,11 +127,24 @@ We evaluated the _roberta-base-ca-cased-tc_ on the TeCla test set against standa
115
 
116
  For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club).
117
 
118
- ## Licensing Information
119
 
 
 
 
 
 
 
 
 
 
 
120
  [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
121
 
122
- ## Citation Information
 
 
 
123
  If you use any of these resources (datasets or models) in your work, please cite our latest paper:
124
  ```bibtex
125
  @inproceedings{armengol-estape-etal-2021-multilingual,
@@ -143,14 +168,6 @@ If you use any of these resources (datasets or models) in your work, please cite
143
  }
144
  ```
145
 
146
- ### Funding
147
- 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).
148
-
149
-
150
- ## Contributions
151
-
152
- [N/A]
153
-
154
  ### Disclaimer
155
 
156
  <details>
 
52
  # Catalan BERTa (roberta-base-ca) finetuned for Text Classification.
53
 
54
  ## Table of Contents
55
+ <details>
56
+ <summary>Click to expand</summary>
57
+
58
+ - [Model description](#model-description)
59
+ - [Intended uses and limitations](#intended-use)
60
+ - [How to use](#how-to-use)
61
+ - [Limitations and bias](#limitations-and-bias)
62
  - [Training](#training)
63
+ - [Training data](#training-data)
64
+ - [Training procedure](#training-procedure)
65
  - [Evaluation](#evaluation)
66
+ - [Variable and metrics](#variable-and-metrics)
67
+ - [Evaluation results](#evaluation-results)
68
+ - [Additional information](#additional-information)
69
+ - [Author](#author)
70
+ - [Contact information](#contact-information)
71
+ - [Copyright](#copyright)
72
+ - [Licensing information](#licensing-information)
73
+ - [Funding](#funding)
74
+ - [Citing information](#citing-information)
75
+ - [Disclaimer](#disclaimer)
76
+ </details>
77
 
78
  ## Model description
79
  The **roberta-base-ca-cased-tc** is a Text Classification (TC) model for the Catalan language fine-tuned from the roberta-base-ca 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.
80
 
81
+ ## Intended uses and limitations
82
 
83
  **roberta-base-ca-cased-tc** model can be used to classify texts. The model is limited by its training dataset and may not generalize well for all use cases.
84
 
85
+ ## How to use
86
 
87
  Here is how to use this model:
88
 
 
97
  pprint(tc_results)
98
  ```
99
 
100
+ ## Limitations and bias
101
+ 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.
102
+
103
+
104
  ## Training
105
 
106
  ### Training data
107
  We used the TC dataset in Catalan called [TeCla](https://huggingface.co/datasets/projecte-aina/tecla) for training and evaluation.
108
 
109
+ ### Training procedure
110
  The model was trained with a batch size of 16 and a learning rate of 5e-5 for 5 epochs. We then selected the best checkpoint using the downstream task metric in the corresponding development set and then evaluated it on the test set.
111
 
112
  ## Evaluation
113
 
114
+ ### Variable and metrics
115
 
116
  This model was finetuned maximizing accuracy.
117
 
 
127
 
128
  For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club).
129
 
130
+ ## Additional information
131
 
132
+ ### Author
133
+ Text Mining Unit (TeMU) at the Barcelona Supercomputing Center (bsc-temu@bsc.es)
134
+
135
+ ### Contact information
136
+ For further information, send an email to aina@bsc.es
137
+
138
+ ### Copyright
139
+ Copyright (c) 2022 Text Mining Unit at Barcelona Supercomputing Center
140
+
141
+ ### Licensing information
142
  [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
143
 
144
+ ### Funding
145
+ 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).
146
+
147
+ ### Citation Information
148
  If you use any of these resources (datasets or models) in your work, please cite our latest paper:
149
  ```bibtex
150
  @inproceedings{armengol-estape-etal-2021-multilingual,
 
168
  }
169
  ```
170
 
 
 
 
 
 
 
 
 
171
  ### Disclaimer
172
 
173
  <details>