update model card README.md

README.md

@@ -24,16 +24,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9394387001477105
+      value: 0.9386094674556213
     - name: Recall
       type: recall
-      value: 0.9520958083832335
+      value: 0.9498502994011976
     - name: F1
       type: f1
-      value: 0.9457249070631969
+      value: 0.9441964285714285
     - name: Accuracy
       type: accuracy
-      value: 0.9550084889643463
+      value: 0.9562818336162988
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,11 +43,11 @@ 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 the cord-layoutlmv3 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2186
-- Precision: 0.9394
-- Recall: 0.9521
-- F1: 0.9457
-- Accuracy: 0.9550
+- Loss: 0.2039
+- Precision: 0.9386
+- Recall: 0.9499
+- F1: 0.9442
+- Accuracy: 0.9563
 
 ## Model description
 
@@ -78,16 +78,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 1.56  | 250  | 1.0245          | 0.7265    | 0.7874 | 0.7557 | 0.7950   |
-| 1.4012        | 3.12  | 500  | 0.5634          | 0.8336    | 0.8660 | 0.8495 | 0.8765   |
-| 1.4012        | 4.69  | 750  | 0.4155          | 0.8823    | 0.9034 | 0.8928 | 0.9143   |
-| 0.4012        | 6.25  | 1000 | 0.3149          | 0.9263    | 0.9319 | 0.9291 | 0.9359   |
-| 0.4012        | 7.81  | 1250 | 0.2733          | 0.9295    | 0.9379 | 0.9337 | 0.9410   |
-| 0.2147        | 9.38  | 1500 | 0.2501          | 0.9319    | 0.9416 | 0.9367 | 0.9448   |
-| 0.2147        | 10.94 | 1750 | 0.2390          | 0.9319    | 0.9424 | 0.9371 | 0.9508   |
-| 0.1472        | 12.5  | 2000 | 0.2231          | 0.9386    | 0.9499 | 0.9442 | 0.9542   |
-| 0.1472        | 14.06 | 2250 | 0.2174          | 0.9408    | 0.9521 | 0.9464 | 0.9563   |
-| 0.1096        | 15.62 | 2500 | 0.2186          | 0.9394    | 0.9521 | 0.9457 | 0.9550   |
+| No log        | 1.56  | 250  | 1.0093          | 0.7425    | 0.7964 | 0.7685 | 0.7984   |
+| 1.3757        | 3.12  | 500  | 0.5393          | 0.8493    | 0.8735 | 0.8613 | 0.8816   |
+| 1.3757        | 4.69  | 750  | 0.3774          | 0.8857    | 0.9049 | 0.8952 | 0.9160   |
+| 0.3755        | 6.25  | 1000 | 0.2909          | 0.9153    | 0.9304 | 0.9228 | 0.9338   |
+| 0.3755        | 7.81  | 1250 | 0.2511          | 0.9174    | 0.9311 | 0.9242 | 0.9393   |
+| 0.1939        | 9.38  | 1500 | 0.2213          | 0.9385    | 0.9484 | 0.9434 | 0.9529   |
+| 0.1939        | 10.94 | 1750 | 0.2176          | 0.9383    | 0.9454 | 0.9418 | 0.9525   |
+| 0.1358        | 12.5  | 2000 | 0.2180          | 0.9314    | 0.9454 | 0.9383 | 0.9503   |
+| 0.1358        | 14.06 | 2250 | 0.2057          | 0.9357    | 0.9484 | 0.9420 | 0.9546   |
+| 0.1035        | 15.62 | 2500 | 0.2039          | 0.9386    | 0.9499 | 0.9442 | 0.9563   |
 
 
 ### Framework versions