ft-ms-layoutlmv3-funsd-layoutlmv3

README.md

@@ -25,16 +25,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.14907749077490776
+      value: 0.89171974522293
     - name: Recall
       type: recall
-      value: 0.20069547938400398
+      value: 0.9041231992051664
     - name: F1
       type: f1
-      value: 0.17107770484861318
+      value: 0.8978786383818451
     - name: Accuracy
       type: accuracy
-      value: 0.40377986449542375
+      value: 0.8377510994888863
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +44,11 @@ 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 funsd-layoutlmv3 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5641
-- Precision: 0.1491
-- Recall: 0.2007
-- F1: 0.1711
-- Accuracy: 0.4038
+- Loss: 1.0021
+- Precision: 0.8917
+- Recall: 0.9041
+- F1: 0.8979
+- Accuracy: 0.8378
 
 ## Model description
 
@@ -68,21 +68,27 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 4
-- eval_batch_size: 4
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- training_steps: 20
+- training_steps: 1000
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 0.13  | 5    | 1.6893          | 0.0938    | 0.1461 | 0.1142 | 0.3520   |
-| No log        | 0.26  | 10   | 1.6149          | 0.1202    | 0.1585 | 0.1367 | 0.3874   |
-| No log        | 0.39  | 15   | 1.5797          | 0.1414    | 0.1913 | 0.1626 | 0.3991   |
-| No log        | 0.53  | 20   | 1.5641          | 0.1491    | 0.2007 | 0.1711 | 0.4038   |
+| No log        | 10.0  | 100  | 0.5222          | 0.8477    | 0.8823 | 0.8647 | 0.8436   |
+| No log        | 20.0  | 200  | 0.6686          | 0.8736    | 0.9026 | 0.8879 | 0.8357   |
+| No log        | 30.0  | 300  | 0.7175          | 0.8759    | 0.9151 | 0.8950 | 0.8286   |
+| No log        | 40.0  | 400  | 0.7636          | 0.8832    | 0.8977 | 0.8904 | 0.8426   |
+| 0.2392        | 50.0  | 500  | 0.9518          | 0.8820    | 0.9026 | 0.8922 | 0.8178   |
+| 0.2392        | 60.0  | 600  | 0.9803          | 0.8771    | 0.8897 | 0.8834 | 0.8121   |
+| 0.2392        | 70.0  | 700  | 1.0956          | 0.8883    | 0.9086 | 0.8983 | 0.8173   |
+| 0.2392        | 80.0  | 800  | 0.9517          | 0.8930    | 0.9076 | 0.9002 | 0.8444   |
+| 0.2392        | 90.0  | 900  | 1.0337          | 0.8950    | 0.9061 | 0.9005 | 0.8379   |
+| 0.0083        | 100.0 | 1000 | 1.0021          | 0.8917    | 0.9041 | 0.8979 | 0.8378   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:677ccd719f843f048663defc035c2cad55d80ae25530dfe736c7fecc9f2e6c0c
+oid sha256:c35c9f77a75a06ecc7ce329e997a9d87004a1aa2a0cf015fbdc3bdbd9208ed1f
 size 501355532

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c79c124ca090eb3267bcb272509d8e8831a3ef29881a26cff32f07bd031e3750
+oid sha256:efd97be0930941513113f2383b55ff36f2942207cea4aacc7988b0b7059b5263
 size 4155