Update model card
Browse files
README.md
CHANGED
@@ -6,12 +6,34 @@ Three classes sentiment analysis (positive, negative, neutral)
|
|
6 |
|
7 |
Based on https://huggingface.co/j-hartmann/sentiment-roberta-large-english-3-classes
|
8 |
|
9 |
-
Fine-tuned using
|
10 |
- annotated sentences from book reviews in English https://www.gti.uvigo.es/index.php/en/book-reviews-annotated-dataset-for-aspect-based-sentiment-analysis
|
11 |
- annotated paragraphs from amateur writers' stories https://arxiv.org/abs/1910.11769
|
12 |
|
13 |
-
Performance for books
|
14 |
-
![hartmann_ft_test_books.png](https://s3.amazonaws.com/moonup/production/uploads/1661430297130-611a6e0f289467cafea62d0e.png)
|
15 |
|
16 |
-
|
17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
|
7 |
Based on https://huggingface.co/j-hartmann/sentiment-roberta-large-english-3-classes
|
8 |
|
9 |
+
**Fine-tuned using:**
|
10 |
- annotated sentences from book reviews in English https://www.gti.uvigo.es/index.php/en/book-reviews-annotated-dataset-for-aspect-based-sentiment-analysis
|
11 |
- annotated paragraphs from amateur writers' stories https://arxiv.org/abs/1910.11769
|
12 |
|
13 |
+
**Performance for books:**
|
|
|
14 |
|
15 |
+
Num examples = 1666
|
16 |
+
|
17 |
+
Batch size = 16
|
18 |
+
| | precision | recall | f1-score | support |
|
19 |
+
|---| --- | --- | --- | --- |
|
20 |
+
| Negative | 0.83 | 0.88 | 0.85 | 844 |
|
21 |
+
| Neutral | 0.68 | 0.51 | 0.58 | 351 |
|
22 |
+
| Positive | 0.79 | 0.85 | 0.82 | 471 |
|
23 |
+
| accuracy | | | 0.79 | 1666 |
|
24 |
+
| macro avg | 0.76 | 0.75 | 0.75 | 1666 |
|
25 |
+
| weighted avg | 0.78 | 0.79 | 0.78 | 1666 |
|
26 |
+
|
27 |
+
**Performance for reviews:**
|
28 |
+
|
29 |
+
Num examples = 205
|
30 |
+
|
31 |
+
Batch size = 16
|
32 |
+
| | precision | recall | f1-score | support |
|
33 |
+
|---| --- | --- | --- | --- |
|
34 |
+
| Negative | 0.89 | 0.92 | 0.91 | 26 |
|
35 |
+
| Neutral | 0.96 | 0.91 | 0.94 | 90 |
|
36 |
+
| Positive | 0.94 | 0.98 | 0.96 | 89 |
|
37 |
+
| accuracy | | | 0.94 | 205 |
|
38 |
+
| macro avg | 0.93 | 0.94 | 0.93 | 205 |
|
39 |
+
| weighted avg | 0.94 | 0.94 | 0.94 | 205 |
|