Update README.md
Browse files
README.md
CHANGED
@@ -48,30 +48,37 @@ widget:
|
|
48 |
# Catalan BERTa-v2 (roberta-base-ca-v2) finetuned for Text Classification.
|
49 |
|
50 |
## Table of Contents
|
51 |
-
|
52 |
-
|
53 |
-
|
|
|
|
|
|
|
|
|
54 |
- [Training](#training)
|
55 |
-
- [Training
|
56 |
-
- [Training
|
57 |
- [Evaluation](#evaluation)
|
58 |
-
- [Variable and
|
59 |
-
- [Evaluation
|
60 |
-
- [
|
61 |
-
- [
|
62 |
-
- [
|
63 |
-
- [
|
64 |
-
- [
|
|
|
|
|
|
|
65 |
|
66 |
## Model description
|
67 |
|
68 |
The **roberta-base-ca-v2-cased-wikicat-ca** is a Text Classification model for the Catalan language fine-tuned from the [roberta-base-ca-v2](https://huggingface.co/projecte-aina/roberta-base-ca-v2) 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 roberta-base-ca-v2 model card for more details).
|
69 |
|
70 |
-
## Intended
|
71 |
|
72 |
**roberta-base-ca-v2-cased-wikicat-ca** model can be used to classify texts. The model is limited by its training dataset and may not generalize well for all use cases.
|
73 |
|
74 |
-
## How to
|
75 |
|
76 |
Here is how to use this model:
|
77 |
|
@@ -86,17 +93,20 @@ tc_results = nlp(example)
|
|
86 |
pprint(tc_results)
|
87 |
```
|
88 |
|
|
|
|
|
|
|
89 |
## Training
|
90 |
|
91 |
### Training data
|
92 |
We used the TC dataset in Catalan called [WikiCAT_ca](https://huggingface.co/datasets/projecte-aina/WikiCAT_ca) for training and evaluation.
|
93 |
|
94 |
-
### Training
|
95 |
The model was trained with a batch size of 16 and three learning rates (1e-5, 3e-5, 5e-5) for 10 epochs. We then selected the best learning rate (3e-5) and checkpoint (epoch 3, step 1857) using the downstream task metric in the corresponding development set.
|
96 |
|
97 |
## Evaluation
|
98 |
|
99 |
-
### Variable and
|
100 |
|
101 |
This model was finetuned maximizing F1 (weighted) score.
|
102 |
|
@@ -109,19 +119,24 @@ We evaluated the _roberta-base-ca-v2-cased-wikicat-ca_ on the WikiCAT_ca dev set
|
|
109 |
|
110 |
For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club).
|
111 |
|
112 |
-
## Licensing Information
|
113 |
|
114 |
-
|
115 |
|
116 |
-
|
|
|
117 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
118 |
|
119 |
### Funding
|
120 |
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).
|
121 |
|
122 |
-
## Contributions
|
123 |
-
|
124 |
-
[N/A]
|
125 |
|
126 |
## Disclaimer
|
127 |
|
|
|
48 |
# Catalan BERTa-v2 (roberta-base-ca-v2) finetuned for Text Classification.
|
49 |
|
50 |
## Table of Contents
|
51 |
+
<details>
|
52 |
+
<summary>Click to expand</summary>
|
53 |
+
|
54 |
+
- [Model description](#model-description)
|
55 |
+
- [Intended uses and limitations](#intended-uses-and-limitations)
|
56 |
+
- [How to use](#how-to-use)
|
57 |
+
- [Limitations and bias](#limitations-and-bias)
|
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 |
+
- [Author](#author)
|
65 |
+
- [Contact information](#contact-information)
|
66 |
+
- [Copyright](#copyright)
|
67 |
+
- [Licensing information](#licensing-information)
|
68 |
+
- [Funding](#funding)
|
69 |
+
- [Disclaimer](#disclaimer)
|
70 |
+
|
71 |
+
</details>
|
72 |
|
73 |
## Model description
|
74 |
|
75 |
The **roberta-base-ca-v2-cased-wikicat-ca** is a Text Classification model for the Catalan language fine-tuned from the [roberta-base-ca-v2](https://huggingface.co/projecte-aina/roberta-base-ca-v2) 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 roberta-base-ca-v2 model card for more details).
|
76 |
|
77 |
+
## Intended uses and limitations
|
78 |
|
79 |
**roberta-base-ca-v2-cased-wikicat-ca** model can be used to classify texts. The model is limited by its training dataset and may not generalize well for all use cases.
|
80 |
|
81 |
+
## How to use
|
82 |
|
83 |
Here is how to use this model:
|
84 |
|
|
|
93 |
pprint(tc_results)
|
94 |
```
|
95 |
|
96 |
+
## Limitations and bias
|
97 |
+
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.
|
98 |
+
|
99 |
## Training
|
100 |
|
101 |
### Training data
|
102 |
We used the TC dataset in Catalan called [WikiCAT_ca](https://huggingface.co/datasets/projecte-aina/WikiCAT_ca) for training and evaluation.
|
103 |
|
104 |
+
### Training procedure
|
105 |
The model was trained with a batch size of 16 and three learning rates (1e-5, 3e-5, 5e-5) for 10 epochs. We then selected the best learning rate (3e-5) and checkpoint (epoch 3, step 1857) using the downstream task metric in the corresponding development set.
|
106 |
|
107 |
## Evaluation
|
108 |
|
109 |
+
### Variable and metrics
|
110 |
|
111 |
This model was finetuned maximizing F1 (weighted) score.
|
112 |
|
|
|
119 |
|
120 |
For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club).
|
121 |
|
|
|
122 |
|
123 |
+
## Additional information
|
124 |
|
125 |
+
### Author
|
126 |
+
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center (bsc-temu@bsc.es)
|
127 |
|
128 |
+
### Contact information
|
129 |
+
For further information, send an email to aina@bsc.es
|
130 |
+
|
131 |
+
### Copyright
|
132 |
+
Copyright by Text Mining Unit - Barcelona Supercomputing Center (2022)
|
133 |
+
|
134 |
+
### Licensing information
|
135 |
+
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
|
136 |
|
137 |
### Funding
|
138 |
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).
|
139 |
|
|
|
|
|
|
|
140 |
|
141 |
## Disclaimer
|
142 |
|