nijatzeynalov
commited on
Commit
•
0727403
1
Parent(s):
da86749
Update README.md
Browse files
README.md
CHANGED
@@ -100,20 +100,20 @@ Token counts are white space based.
|
|
100 |
|
101 |
## Training results with comparison
|
102 |
|
103 |
-
|
104 |
|
105 |
-
Rouge1: 39.4222
|
106 |
-
Rouge2: 24.8624
|
107 |
-
Rougel: 32.2487
|
108 |
|
109 |
For __Azerbaijani text summarization downstream task__, mT5-multilingual-XLSum has also been developed on the 45 languages of [XL-Sum](https://huggingface.co/datasets/csebuetnlp/xlsum) dataset. For finetuning details and scripts,
|
110 |
see the [paper](https://aclanthology.org/2021.findings-acl.413/) and the [official repository](https://github.com/csebuetnlp/xl-sum). .
|
111 |
|
112 |
-
|
113 |
|
114 |
-
Rouge1: 21.4227
|
115 |
-
Rouge2: 9.5214
|
116 |
-
Rougel: 19.3331
|
117 |
|
118 |
As seen from the numbers, __mT5-based-azerbaijani-summarize__ model achieves dramatically better performance than __mT5_multilingual_XLSum__.
|
119 |
|
|
|
100 |
|
101 |
## Training results with comparison
|
102 |
|
103 |
+
__mT5-based-azerbaijani-summarize__ model rouge scores on the test set:
|
104 |
|
105 |
+
- Rouge1: 39.4222
|
106 |
+
- Rouge2: 24.8624
|
107 |
+
- Rougel: 32.2487
|
108 |
|
109 |
For __Azerbaijani text summarization downstream task__, mT5-multilingual-XLSum has also been developed on the 45 languages of [XL-Sum](https://huggingface.co/datasets/csebuetnlp/xlsum) dataset. For finetuning details and scripts,
|
110 |
see the [paper](https://aclanthology.org/2021.findings-acl.413/) and the [official repository](https://github.com/csebuetnlp/xl-sum). .
|
111 |
|
112 |
+
__mT5_multilingual_XLSum__ modelrouge scores on the XL-Sum test set (only for Azerbaijani):
|
113 |
|
114 |
+
- Rouge1: 21.4227
|
115 |
+
- Rouge2: 9.5214
|
116 |
+
- Rougel: 19.3331
|
117 |
|
118 |
As seen from the numbers, __mT5-based-azerbaijani-summarize__ model achieves dramatically better performance than __mT5_multilingual_XLSum__.
|
119 |
|