Added citations
Browse files
README.md
CHANGED
@@ -9,7 +9,7 @@ widget:
|
|
9 |
- text: "Silva, ti si grda in neprijazna"
|
10 |
---
|
11 |
|
12 |
-
Text classification model based on `EMBEDDIA/sloberta` and fine-tuned on the [
|
13 |
|
14 |
## Fine-tuning hyperparameters
|
15 |
|
@@ -69,4 +69,47 @@ predictions, logit_output = model.predict(["Silva, ti si grda in neprijazna", "N
|
|
69 |
predictions
|
70 |
### Output:
|
71 |
### array([1, 0])
|
72 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
- text: "Silva, ti si grda in neprijazna"
|
10 |
---
|
11 |
|
12 |
+
Text classification model based on `EMBEDDIA/sloberta` and fine-tuned on the [FRENK dataset](https://www.clarin.si/repository/xmlui/handle/11356/1433) comprising of LGBT and migrant hatespeech. Only the slovenian subset of the data was used for fine-tuning and the dataset has been relabeled for binary classification (offensive or acceptable).
|
13 |
|
14 |
## Fine-tuning hyperparameters
|
15 |
|
|
|
69 |
predictions
|
70 |
### Output:
|
71 |
### array([1, 0])
|
72 |
+
```
|
73 |
+
|
74 |
+
## Citation
|
75 |
+
|
76 |
+
If you use the model, please cite the following paper on which the original model is based:
|
77 |
+
|
78 |
+
```
|
79 |
+
@article{DBLP:journals/corr/abs-1907-11692,
|
80 |
+
author = {Yinhan Liu and
|
81 |
+
Myle Ott and
|
82 |
+
Naman Goyal and
|
83 |
+
Jingfei Du and
|
84 |
+
Mandar Joshi and
|
85 |
+
Danqi Chen and
|
86 |
+
Omer Levy and
|
87 |
+
Mike Lewis and
|
88 |
+
Luke Zettlemoyer and
|
89 |
+
Veselin Stoyanov},
|
90 |
+
title = {RoBERTa: {A} Robustly Optimized {BERT} Pretraining Approach},
|
91 |
+
journal = {CoRR},
|
92 |
+
volume = {abs/1907.11692},
|
93 |
+
year = {2019},
|
94 |
+
url = {http://arxiv.org/abs/1907.11692},
|
95 |
+
archivePrefix = {arXiv},
|
96 |
+
eprint = {1907.11692},
|
97 |
+
timestamp = {Thu, 01 Aug 2019 08:59:33 +0200},
|
98 |
+
biburl = {https://dblp.org/rec/journals/corr/abs-1907-11692.bib},
|
99 |
+
bibsource = {dblp computer science bibliography, https://dblp.org}
|
100 |
+
}
|
101 |
+
|
102 |
+
```
|
103 |
+
|
104 |
+
and the dataset used for fine-tuning:
|
105 |
+
|
106 |
+
```
|
107 |
+
@misc{ljubešić2019frenk,
|
108 |
+
title={The FRENK Datasets of Socially Unacceptable Discourse in Slovene and English},
|
109 |
+
author={Nikola Ljubešić and Darja Fišer and Tomaž Erjavec},
|
110 |
+
year={2019},
|
111 |
+
eprint={1906.02045},
|
112 |
+
archivePrefix={arXiv},
|
113 |
+
primaryClass={cs.CL},
|
114 |
+
url={https://arxiv.org/abs/1906.02045}
|
115 |
+
}
|