Niyazi's picture
update model card README.md
94fb79a
metadata
license: cc-by-nc-sa-4.0
tags:
  - generated_from_trainer
datasets:
  - cord-layoutlmv3
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: layoutlmv3-finetuned-cord_100
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: cord-layoutlmv3
          type: cord-layoutlmv3
          config: cord
          split: test
          args: cord
        metrics:
          - name: Precision
            type: precision
            value: 0.9479940564635958
          - name: Recall
            type: recall
            value: 0.9550898203592815
          - name: F1
            type: f1
            value: 0.9515287099179717
          - name: Accuracy
            type: accuracy
            value: 0.9562818336162988

layoutlmv3-finetuned-cord_100

This model is a fine-tuned version of microsoft/layoutlmv3-base on the cord-layoutlmv3 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2163
  • Precision: 0.9480
  • Recall: 0.9551
  • F1: 0.9515
  • Accuracy: 0.9563

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 10
  • eval_batch_size: 10
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 2500

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 3.12 250 0.8807 0.7910 0.8301 0.8101 0.8247
1.2597 6.25 500 0.4375 0.8808 0.9072 0.8938 0.9117
1.2597 9.38 750 0.3119 0.9185 0.9364 0.9274 0.9423
0.2869 12.5 1000 0.2700 0.9340 0.9424 0.9382 0.9452
0.2869 15.62 1250 0.2401 0.9429 0.9513 0.9471 0.9559
0.1378 18.75 1500 0.2295 0.9415 0.9521 0.9468 0.9550
0.1378 21.88 1750 0.2183 0.9523 0.9566 0.9544 0.9580
0.0866 25.0 2000 0.2190 0.9523 0.9566 0.9544 0.9576
0.0866 28.12 2250 0.2168 0.9508 0.9551 0.9529 0.9567
0.066 31.25 2500 0.2163 0.9480 0.9551 0.9515 0.9563

Framework versions

  • Transformers 4.26.1
  • Pytorch 1.13.0
  • Datasets 2.10.1
  • Tokenizers 0.13.2