Update README.md

README.md

@@ -20,7 +20,7 @@ tags:
 *wsd-camembert-base-semcor-wngt-fr* is a Word Sense Disambiguation (WSD) model fine-tuned on the French version of Semcor and WNGT datasets with *almanach/camembert-base* as the pretrained BERT embeddings.
 
 The fine-tuned model achieves the following performance on SemEval 2013 - fr:
-| Test F1 (%) | GPUs | Epochs
+| Test F1 (%) | GPUs | Epochs |
 |:-------------:|:--------------:|:--------------:| :--------:|:--------:|
 | 51.28 | 1xV100 32GB | 40 |
 
@@ -36,11 +36,10 @@ To disambiguate a sentence, please refer to the official [NWSD](https://github.c
 
 ### Training and Test Data
 
-We use [Semcor.fr](https://frsemcor.github.io/FrSemCor/) and [WNGT.fr](https://github.com/getalp/UFSAC) annotated with WordNet 3.0 sense keys IDs with:
+We use [Semcor.fr](https://frsemcor.github.io/FrSemCor/) and [WNGT.fr](https://github.com/getalp/UFSAC) annotated with WordNet 3.0 sense keys IDs for the train/valid sets:
 | | Train | Valid |
 |:-------------:|:-------------:|:--------------:|
-| # utterances | 143,597 | 4,000 |
-
+| # utterances | 143,597 | 4,000 | 306
 The semeval2013task12.fr.xml test data is the French version of the [SemEval-2013 Task 12](https://aclanthology.org/S13-2040/) test set, with:
 | | Test | 
 |:-------------:|:-------------:|