mp-02
/

layoutlmv3-finetuned-cord

@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1816
 - Precision: 0.9617
 - Recall: 0.9752
 - F1: 0.9684
-- Accuracy: 0.9686
 ## Model description
@@ -54,18 +54,23 @@ The following hyperparameters were used during training:
 ### Training results
-| Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
-|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 1.3286        | 1.875  | 150  | 0.4894          | 0.8378    | 0.8905 | 0.8634 | 0.8829   |
-| 0.3358        | 3.75   | 300  | 0.2462          | 0.9264    | 0.9573 | 0.9416 | 0.9508   |
-| 0.1629        | 5.625  | 450  | 0.1974          | 0.9422    | 0.9620 | 0.9520 | 0.9588   |
-| 0.0948        | 7.5    | 600  | 0.1913          | 0.9534    | 0.9689 | 0.9611 | 0.9618   |
-| 0.058         | 9.375  | 750  | 0.1741          | 0.9543    | 0.9720 | 0.9631 | 0.9660   |
-| 0.0409        | 11.25  | 900  | 0.1738          | 0.9573    | 0.9759 | 0.9666 | 0.9694   |
-| 0.0285        | 13.125 | 1050 | 0.1981          | 0.9571    | 0.9697 | 0.9634 | 0.9635   |
-| 0.0256        | 15.0   | 1200 | 0.1863          | 0.9580    | 0.9728 | 0.9653 | 0.9665   |
-| 0.0192        | 16.875 | 1350 | 0.1802          | 0.9640    | 0.9767 | 0.9703 | 0.9703   |
-| 0.0143        | 18.75  | 1500 | 0.1816          | 0.9617    | 0.9752 | 0.9684 | 0.9686   |
 ### Framework versions

 This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.1756
 - Precision: 0.9617
 - Recall: 0.9752
 - F1: 0.9684
+- Accuracy: 0.9707
 ## Model description
 ### Training results
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.8575        | 1.25  | 100  | 0.7617          | 0.7943    | 0.8455 | 0.8191 | 0.8345   |
+| 0.411         | 2.5   | 200  | 0.3537          | 0.8834    | 0.9293 | 0.9058 | 0.9198   |
+| 0.2338        | 3.75  | 300  | 0.2471          | 0.9396    | 0.9666 | 0.9529 | 0.9546   |
+| 0.175         | 5.0   | 400  | 0.1970          | 0.9460    | 0.9666 | 0.9562 | 0.9605   |
+| 0.0761        | 6.25  | 500  | 0.1824          | 0.9548    | 0.9674 | 0.9610 | 0.9648   |
+| 0.07          | 7.5   | 600  | 0.1839          | 0.9498    | 0.9697 | 0.9597 | 0.9660   |
+| 0.1061        | 8.75  | 700  | 0.1685          | 0.9616    | 0.9728 | 0.9672 | 0.9699   |
+| 0.0197        | 10.0  | 800  | 0.1704          | 0.9586    | 0.9697 | 0.9641 | 0.9665   |
+| 0.0238        | 11.25 | 900  | 0.1793          | 0.9608    | 0.9705 | 0.9656 | 0.9656   |
+| 0.0149        | 12.5  | 1000 | 0.2080          | 0.9654    | 0.9744 | 0.9699 | 0.9652   |
+| 0.0208        | 13.75 | 1100 | 0.1708          | 0.9602    | 0.9728 | 0.9664 | 0.9699   |
+| 0.0253        | 15.0  | 1200 | 0.1783          | 0.9617    | 0.9744 | 0.9680 | 0.9699   |
+| 0.0213        | 16.25 | 1300 | 0.1754          | 0.9655    | 0.9767 | 0.9711 | 0.9724   |
+| 0.0079        | 17.5  | 1400 | 0.1715          | 0.9670    | 0.9775 | 0.9722 | 0.9728   |
+| 0.0226        | 18.75 | 1500 | 0.1756          | 0.9617    | 0.9752 | 0.9684 | 0.9707   |
 ### Framework versions