jjmcarrascosa
commited on
Commit
•
5d23d47
1
Parent(s):
d23dbb1
Update README.md
Browse files
README.md
CHANGED
@@ -19,7 +19,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
|
20 |
# xlm-roberta-base-finetuned
|
21 |
|
22 |
-
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on
|
|
|
|
|
|
|
|
|
|
|
23 |
It achieves the following results on the evaluation set:
|
24 |
- eval_loss: 0.4852
|
25 |
- eval_f1: 0.8009
|
|
|
19 |
|
20 |
# xlm-roberta-base-finetuned
|
21 |
|
22 |
+
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on 2 datasets, labelled with not_toxic (0) / toxic (1) content from news or tweets.
|
23 |
+
- a private one, provided by @Newtral, containing both tweets and news.
|
24 |
+
- one used for data augmentation purposes, containing only news, obtained from [SurgeHQ.ai](https://app.surgehq.ai/datasets/spanish-toxicity)
|
25 |
+
|
26 |
+
The test dataset was provided by @Newtral and was kept fixed.
|
27 |
+
|
28 |
It achieves the following results on the evaluation set:
|
29 |
- eval_loss: 0.4852
|
30 |
- eval_f1: 0.8009
|