Update README.md
Browse files
README.md
CHANGED
@@ -33,7 +33,8 @@ metrics:
|
|
33 |
|
34 |
## Model description
|
35 |
|
36 |
-
This model was trained from scratch using the [Fairseq toolkit](https://fairseq.readthedocs.io/en/latest/) on a combination of Catalan-Portuguese datasets,
|
|
|
37 |
|
38 |
## Intended uses and limitations
|
39 |
|
@@ -63,6 +64,11 @@ translated = translator.translate_batch([tokenized[0]])
|
|
63 |
print(tokenizer.detokenize(translated[0][0]['tokens']))
|
64 |
```
|
65 |
|
|
|
|
|
|
|
|
|
|
|
66 |
## Training
|
67 |
|
68 |
### Training data
|
@@ -133,11 +139,12 @@ The model was trained for a total of 17.000 updates. Weights were saved every 10
|
|
133 |
|
134 |
### Variable and metrics
|
135 |
|
136 |
-
We use the BLEU score for evaluation on the [Flores-101](https://github.com/facebookresearch/flores) and [NTREX](https://github.com/MicrosoftTranslator/NTREX) test sets
|
137 |
|
138 |
### Evaluation results
|
139 |
|
140 |
-
Below are the evaluation results on the machine translation from Catalan to Portuguese
|
|
|
141 |
|
142 |
| Test set | SoftCatalà | Google Translate |mt-aina-ca-pt|
|
143 |
|----------------------|------------|------------------|---------------|
|
@@ -149,29 +156,34 @@ Below are the evaluation results on the machine translation from Catalan to Port
|
|
149 |
## Additional information
|
150 |
|
151 |
### Author
|
152 |
-
Language Technologies Unit
|
153 |
|
154 |
-
### Contact
|
155 |
-
For further information, please send an email to langtech@bsc.es
|
156 |
|
157 |
### Copyright
|
158 |
-
Copyright Language Technologies Unit
|
159 |
|
160 |
-
###
|
161 |
-
|
162 |
|
163 |
### Funding
|
164 |
-
This work
|
165 |
-
|
166 |
-
## Limitations and Bias
|
167 |
-
At the time of submission, no measures have been taken to estimate the bias and toxicity embedded in the model. However, we are 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.
|
168 |
|
169 |
### Disclaimer
|
170 |
|
171 |
<details>
|
172 |
<summary>Click to expand</summary>
|
173 |
|
174 |
-
The
|
175 |
-
|
176 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
177 |
</details>
|
|
|
33 |
|
34 |
## Model description
|
35 |
|
36 |
+
This model was trained from scratch using the [Fairseq toolkit](https://fairseq.readthedocs.io/en/latest/) on a combination of Catalan-Portuguese datasets,
|
37 |
+
which after filtering and cleaning comprised 6.159.631 sentence pairs. The model was evaluated on the Flores and NTREX evaluation datasets.
|
38 |
|
39 |
## Intended uses and limitations
|
40 |
|
|
|
64 |
print(tokenizer.detokenize(translated[0][0]['tokens']))
|
65 |
```
|
66 |
|
67 |
+
## Limitations and bias
|
68 |
+
At the time of submission, no measures have been taken to estimate the bias and toxicity embedded in the model.
|
69 |
+
However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques
|
70 |
+
on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.
|
71 |
+
|
72 |
## Training
|
73 |
|
74 |
### Training data
|
|
|
139 |
|
140 |
### Variable and metrics
|
141 |
|
142 |
+
We use the BLEU score for evaluation on the [Flores-101](https://github.com/facebookresearch/flores) and [NTREX](https://github.com/MicrosoftTranslator/NTREX) test sets.
|
143 |
|
144 |
### Evaluation results
|
145 |
|
146 |
+
Below are the evaluation results on the machine translation from Catalan to Portuguese
|
147 |
+
compared to [Softcatalà](https://www.softcatala.org/) and [Google Translate](https://translate.google.es/?hl=es):
|
148 |
|
149 |
| Test set | SoftCatalà | Google Translate |mt-aina-ca-pt|
|
150 |
|----------------------|------------|------------------|---------------|
|
|
|
156 |
## Additional information
|
157 |
|
158 |
### Author
|
159 |
+
The Language Technologies Unit from Barcelona Supercomputing Center.
|
160 |
|
161 |
+
### Contact
|
162 |
+
For further information, please send an email to <langtech@bsc.es>.
|
163 |
|
164 |
### Copyright
|
165 |
+
Copyright(c) 2023 by Language Technologies Unit, Barcelona Supercomputing Center.
|
166 |
|
167 |
+
### License
|
168 |
+
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
|
169 |
|
170 |
### Funding
|
171 |
+
This work was funded by [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).
|
|
|
|
|
|
|
172 |
|
173 |
### Disclaimer
|
174 |
|
175 |
<details>
|
176 |
<summary>Click to expand</summary>
|
177 |
|
178 |
+
The model published in this repository is intended for a generalist purpose and is available to third parties under a permissive Apache License, Version 2.0.
|
179 |
+
|
180 |
+
Be aware that the model may have biases and/or any other undesirable distortions.
|
181 |
+
|
182 |
+
When third parties deploy or provide systems and/or services to other parties using this model (or any system based on it)
|
183 |
+
or become users of the model, they should note that it is their responsibility to mitigate the risks arising from its use and,
|
184 |
+
in any event, to comply with applicable regulations, including regulations regarding the use of Artificial Intelligence.
|
185 |
+
|
186 |
+
In no event shall the owner and creator of the model (Barcelona Supercomputing Center)
|
187 |
+
be liable for any results arising from the use made by third parties.
|
188 |
+
|
189 |
</details>
|