Update README.md

README.md

@@ -24,7 +24,7 @@ model-index:
       name: openkp
     metrics:
       - type: seqeval 
-        value: 0.0
+        value: 0.430
         name: F1-score
 ---
 ** Work in progress **
@@ -112,9 +112,9 @@ print(keyphrases)
 ```
 
 ## 📚 Training Dataset
-OpenKP is a keyphrase extraction/generation dataset consisting of 
+OpenKP is a large-scale, open-domain keyphrase extraction dataset with 148,124 real-world web documents along with 1-3 most relevant human-annotated keyphrases.
 
-You can find more information here: https://huggingface.co/datasets/midas/openkp
+You can find more information here: https://github.com/microsoft/OpenKP
 
 ## 👷‍♂️ Training procedure
 For more in detail information, you can take a look at the training notebook (link incoming).
@@ -205,7 +205,7 @@ The model achieves the following results on the OpenKP test set:
 
 | Dataset           | P@5  | R@5  | F1@5 | P@10 | R@10 | F1@10 | P@M  | R@M  | F1@M |
 |:-----------------:|:----:|:----:|:----:|:----:|:----:|:-----:|:----:|:----:|:----:|
-| OpenKP Test Set   | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00  | 0.00 | 0.00 | 0.00 |
+| OpenKP Test Set   | 0.12 | 0.33 | 0.17 | 0.06 | 0.33 | 0.10  | 0.35 | 0.33 | 0.31 |
 
 For more information on the evaluation process, you can take a look at the keyphrase extraction evaluation notebook.