Update README.md

README.md

@@ -1,8 +1,6 @@
 ---
 tags:
 - generated_from_trainer
-datasets:
-- rubric
 metrics:
 - accuracy
 model-index:
@@ -11,10 +9,6 @@ model-index:
   - task:
       name: Text Classification
       type: text-classification
-    dataset:
-      name: rubric
-      type: rubric
-      args: rubric
     metrics:
     - name: Accuracy
       type: accuracy
@@ -26,8 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # EstBERT128_Rubric
 
-This model is a fine-tuned version of [tartuNLP/EstBERT](https://huggingface.co/tartuNLP/EstBERT) on the rubric categories of the [Estonian Valence dataset](http://peeter.eki.ee:5000/valence/paragraphsquery).
-The data was split into train/dev/test parts with 70/10/20 proportions.
+This model is a fine-tuned version of [tartuNLP/EstBERT](https://huggingface.co/tartuNLP/EstBERT).
 It achieves the following results on the test set:
 - Loss: 2.0552
 - Accuracy: 0.8329
@@ -38,7 +31,16 @@ A single linear layer classifier is fit on top of the last layer [CLS] token rep
 
 ## Intended uses & limitations
 
-This model is intended to be used as it is. It can be used to predict nine rubric categories of Estonian texts. The nine rubric labels in the Estonian Valence dataset are:
+This model is intended to be used as it is. It can be used to predict nine rubric categories of Estonian texts. 
+
+We do not guarantee that the model is useful for anything or that the predictions are accurate on new data.
+
+## Training and evaluation data
+
+The model was trained and evaluated on the rubric categories of the [Estonian Valence dataset](http://peeter.eki.ee:5000/valence/paragraphsquery).
+The data was split into train/dev/test parts with 70/10/20 proportions.
+
+The nine rubric labels in the Estonian Valence dataset are:
 - ARVAMUS (opinion)
 - EESTI (domestic)
 - ELU-O (life)
@@ -51,13 +53,9 @@ This model is intended to be used as it is. It can be used to predict nine rubri
 
 It probably makes sense to treat the two comments categories (KOMM-O-ELU and KOMM-P-EESTI) as a single category.
 
-We do not guarantee that the model is useful for anything or that the predictions are accurate on new data.
-
-## Training and evaluation data
-
-More information needed
-
 ## Training procedure
+The model was trained for maximu 100 epochs using early stopping procedure. After every epoch, the accuracy was calculated on the development set. 
+If the development set accuracy did not improve for 20 epochs, the training was stopped.
 
 ### Training hyperparameters
 
@@ -72,6 +70,7 @@ The following hyperparameters were used during training:
 - mixed_precision_training: Native AMP
 
 ### Training results
+The final model was taken after 39th epoch.
 
 | Training Loss | Epoch | Step  | Validation Loss | Accuracy |
 |:-------------:|:-----:|:-----:|:---------------:|:--------:|