Francesca Del Bonifro commited on
Commit ·
c0688bc
1
Parent(s): b21080b
Update README.md
Browse filesfix labels meaning
README.md
CHANGED
|
@@ -5,8 +5,9 @@ license: apache-2.0
|
|
| 5 |
**I**talian-**S**entence **P**air **In**ference, AKA **I-SPIn**.<br>
|
| 6 |
This is a fine-tuned version of the model [paraphrase-multilingual-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2).<br>
|
| 7 |
Its main task is to perform the [Natural Language Inference (NLI)](https://nlp.stanford.edu/projects/snli/) task in the Italian language.<br>
|
| 8 |
-
The prediction label has two possible values: 1 means the model predicts <em>entailment</em>,
|
| 9 |
-
|
|
|
|
| 10 |
## How it was trained
|
| 11 |
1. Train [paraphrase-multilingual-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2) on the NLI task;
|
| 12 |
2. Apply Knowledge Distillation on the output of (1.) with IT-EN translation dataset to retain NLI knowledge and improve Italian language comprehension.
|
|
|
|
| 5 |
**I**talian-**S**entence **P**air **In**ference, AKA **I-SPIn**.<br>
|
| 6 |
This is a fine-tuned version of the model [paraphrase-multilingual-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2).<br>
|
| 7 |
Its main task is to perform the [Natural Language Inference (NLI)](https://nlp.stanford.edu/projects/snli/) task in the Italian language.<br>
|
| 8 |
+
The prediction label has two possible values: 1 means the model predicts <em>entailment</em>, 0 represents the <em>neutral</em> case
|
| 9 |
+
while -1 corresponds to <em>contradiction</em>.
|
| 10 |
+
|
| 11 |
## How it was trained
|
| 12 |
1. Train [paraphrase-multilingual-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2) on the NLI task;
|
| 13 |
2. Apply Knowledge Distillation on the output of (1.) with IT-EN translation dataset to retain NLI knowledge and improve Italian language comprehension.
|