Training complete

README.md

@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [konverner/distilcamembert-base-ner-address](https://huggingface.co/konverner/distilcamembert-base-ner-address) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0918
-- Precision: 0.9412
+- Loss: 0.1194
+- Precision: 0.9268
 - Recall: 1.0
-- F1: 0.9697
-- Accuracy: 0.9821
+- F1: 0.9620
+- Accuracy: 0.9848
 
 ## Model description
 
@@ -54,9 +54,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 1.0   | 33   | 0.1685          | 0.7949    | 0.9688 | 0.8732 | 0.9430   |
-| No log        | 2.0   | 66   | 0.0946          | 0.9412    | 1.0    | 0.9697 | 0.9800   |
-| No log        | 3.0   | 99   | 0.0918          | 0.9412    | 1.0    | 0.9697 | 0.9821   |
+| No log        | 1.0   | 64   | 0.1575          | 0.7778    | 0.9211 | 0.8434 | 0.9727   |
+| No log        | 2.0   | 128  | 0.1508          | 0.95      | 1.0    | 0.9744 | 0.9783   |
+| No log        | 3.0   | 192  | 0.1194          | 0.9268    | 1.0    | 0.9620 | 0.9848   |
 
 
 ### Framework versions