nitesh-sawadia's picture
update model card README.md
247eee4
metadata
license: cc-by-nc-sa-4.0
tags:
  - generated_from_trainer
datasets:
  - invoice_layoutlmv3
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: layoutlmv3-finetuned-intellectai
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: invoice_layoutlmv3
          type: invoice_layoutlmv3
          config: intellectai
          split: validation
          args: intellectai
        metrics:
          - name: Precision
            type: precision
            value: 0.7053571428571429
          - name: Recall
            type: recall
            value: 0.8540540540540541
          - name: F1
            type: f1
            value: 0.7726161369193154
          - name: Accuracy
            type: accuracy
            value: 0.9624772313296903

layoutlmv3-finetuned-intellectai

This model is a fine-tuned version of microsoft/layoutlmv3-base on the invoice_layoutlmv3 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3645
  • Precision: 0.7054
  • Recall: 0.8541
  • F1: 0.7726
  • Accuracy: 0.9625

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: 2
  • eval_batch_size: 2
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 500

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 0.79 50 1.7979 0.0228 0.0541 0.0321 0.1410
No log 1.59 100 1.2400 0.0863 0.4216 0.1433 0.2616
No log 2.38 150 0.8691 0.1279 0.6919 0.2159 0.4495
No log 3.17 200 0.6001 0.2323 0.8162 0.3617 0.7570
No log 3.97 250 0.4709 0.4660 0.7784 0.5830 0.9093
No log 4.76 300 0.3986 0.5977 0.8270 0.6939 0.9472
No log 5.56 350 0.3762 0.5714 0.8216 0.6741 0.9454
No log 6.35 400 0.3763 0.7048 0.8649 0.7767 0.9636
No log 7.14 450 0.3696 0.6639 0.8541 0.7470 0.9570
0.71 7.94 500 0.3645 0.7054 0.8541 0.7726 0.9625

Framework versions

  • Transformers 4.27.1
  • Pytorch 1.13.1+cu116
  • Datasets 2.10.1
  • Tokenizers 0.13.2