Tommert25
/

robbert0410_lrate2.5b4

Token Classification

Transformers

PyTorch

roberta

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Community

Tommert25 commited on Oct 4, 2023

Commit

f63074e

1 Parent(s): 18fdde5

End of training

Browse files

Files changed (1) hide show

README.md +14 -14

README.md CHANGED Viewed

@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [pdelobelle/robbert-v2-dutch-base](https://huggingface.co/pdelobelle/robbert-v2-dutch-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6503
-- Precisions: 0.8243
-- Recall: 0.7775
-- F-measure: 0.7978
-- Accuracy: 0.9076
 ## Model description
@@ -41,7 +41,7 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
 - train_batch_size: 4
 - eval_batch_size: 4
 - seed: 42
@@ -53,14 +53,14 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:----------:|:------:|:---------:|:--------:|
-| 0.6524        | 1.0   | 942  | 0.4534          | 0.8361     | 0.6745 | 0.6791    | 0.8712   |
-| 0.372         | 2.0   | 1884 | 0.4010          | 0.7997     | 0.7323 | 0.7446    | 0.8930   |
-| 0.23          | 3.0   | 2826 | 0.4862          | 0.7950     | 0.7439 | 0.7519    | 0.8963   |
-| 0.1446        | 4.0   | 3768 | 0.5515          | 0.8251     | 0.7729 | 0.7851    | 0.9080   |
-| 0.0916        | 5.0   | 4710 | 0.6213          | 0.7940     | 0.7793 | 0.7849    | 0.9036   |
-| 0.0597        | 6.0   | 5652 | 0.6503          | 0.8243     | 0.7775 | 0.7978    | 0.9076   |
-| 0.0317        | 7.0   | 6594 | 0.6706          | 0.8098     | 0.7824 | 0.7922    | 0.9107   |
-| 0.021         | 8.0   | 7536 | 0.6690          | 0.8172     | 0.7847 | 0.7968    | 0.9135   |
 ### Framework versions

 This model is a fine-tuned version of [pdelobelle/robbert-v2-dutch-base](https://huggingface.co/pdelobelle/robbert-v2-dutch-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.6518
+- Precisions: 0.8163
+- Recall: 0.7936
+- F-measure: 0.8017
+- Accuracy: 0.9116
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 2.5e-05
 - train_batch_size: 4
 - eval_batch_size: 4
 - seed: 42
 | Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:----------:|:------:|:---------:|:--------:|
+| 0.62          | 1.0   | 942  | 0.4300          | 0.8644     | 0.6922 | 0.7030    | 0.8830   |
+| 0.3475        | 2.0   | 1884 | 0.4044          | 0.8222     | 0.7322 | 0.7464    | 0.8970   |
+| 0.2227        | 3.0   | 2826 | 0.4658          | 0.7715     | 0.7573 | 0.7476    | 0.9070   |
+| 0.1488        | 4.0   | 3768 | 0.5292          | 0.8193     | 0.7461 | 0.7655    | 0.9045   |
+| 0.0983        | 5.0   | 4710 | 0.5855          | 0.7938     | 0.7749 | 0.7829    | 0.9049   |
+| 0.0652        | 6.0   | 5652 | 0.6155          | 0.8170     | 0.7826 | 0.7976    | 0.9100   |
+| 0.0419        | 7.0   | 6594 | 0.6306          | 0.8072     | 0.7929 | 0.7971    | 0.9123   |
+| 0.032         | 8.0   | 7536 | 0.6518          | 0.8163     | 0.7936 | 0.8017    | 0.9116   |
 ### Framework versions