Update README.md

README.md

@@ -14,6 +14,7 @@ widget:
 - text: "Azitromicina en suspensión oral, 10 mg/kg una vez al día durante siete días"
 - text: "A un grupo se le administró Ciprofloxacino 200 mg bid EV y al otro Cefazolina 1 g tid IV"
 - text: "Administración de una solución de mantenimiento intravenosa isotónica (NaCl al 0,9% en dextrosa al 5%)"
+- text: "Se excluyen pacientes con contraindicación a aspirina o clopidogrel"
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -27,9 +28,9 @@ This named entity recognition model detects medication attributes:
 - Route: e.g. *vía oral*, *i.v.*
 
 The model achieves the following results on the test set (when trained with the training and development set; results are averaged over 5 evaluation rounds):
-- Precision: 0.853 (±0.008)
-- Recall: 0.890 (±0.014)
-- F1: 0.871 (±0.008)
+- Precision: 0.863 (±0.011)
+- Recall: 0.878 (±0.008)
+- F1: 0.871 (±0.001)
 - Accuracy: 0.997 (±0.001)
 
 
@@ -93,23 +94,24 @@ The following hyperparameters were used during training:
 - seed: we used different seeds for 5 evaluation rounds, and uploaded the model with the best results
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 8 
+- num_epochs: average 12 epochs (±3.1); trained with early stopping if no improvement after 5 epochs (early stopping patience: 5) 
 
 
 ### Training results (test set; average and standard deviation of 5 rounds with different seeds)
 
 |   Precision    |     Recall     |       F1       |    Accuracy    |
 |:--------------:|:--------------:|:--------------:|:--------------:|
-| 0.853 (±0.008) | 0.890 (±0.014) | 0.871 (±0.008) | 0.997 (±0.001) |
+| 0.863 (±0.011) | 0.878 (±0.008) | 0.871 (±0.001) | 0.997 (±0.001) |
 
 
 **Results per class (test set; average and standard deviation of 5 rounds with different seeds)**  
 
-| Class      |   Precision    |     Recall     |       F1       |  Support  |
-|:----------:|:--------------:|:--------------:|:--------------:|:---------:|
-| Dose       | 0.812 (±0.030) | 0.873 (±0.022) | 0.841 (±0.014) |    314    |
-| Form       | 0.818 (±0.009) | 0.883 (±0.023) | 0.849 (±0.013) |     65    |
-| Route      | 0.916 (±0.032) | 0.911 (±0.012) | 0.913 (±0.019) |    270    |
+|      Class      |   Precision    |     Recall     |       F1       |  Support  |
+|:---------------:|:--------------:|:--------------:|:--------------:|:---------:|
+| Contraindicated | 0.752 (±0.089) | 0.847 (±0.077) | 0.791 (±0.041) |     76    |
+| Dose            | 0.830 (±0.032) | 0.838 (±0.035) | 0.833 (±0.001) |    320    |
+| Form            | 0.971 (±0.029) | 0.889 (±0.024) | 0.928 (±0.021) |     74    |
+| Route           | 0.934 (±0.012) | 0.916 (±0.024) | 0.925 (±0.012) |    270    |
 
 
 ### Framework versions