Edit model card

layoutlmv3-finetuned-cord_800

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: 0.2042
  • Precision: 0.9445
  • Recall: 0.9558
  • F1: 0.9501
  • Accuracy: 0.9605

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: 5
  • eval_batch_size: 5
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 4000

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.56 250 0.9737 0.7787 0.8166 0.7972 0.8188
1.3706 3.12 500 0.5489 0.8480 0.8645 0.8562 0.8680
1.3706 4.69 750 0.3857 0.8913 0.9087 0.8999 0.9147
0.3693 6.25 1000 0.3192 0.9117 0.9274 0.9195 0.9317
0.3693 7.81 1250 0.2816 0.9189 0.9326 0.9257 0.9355
0.1903 9.38 1500 0.2521 0.9277 0.9409 0.9342 0.9465
0.1903 10.94 1750 0.2353 0.9357 0.9476 0.9416 0.9550
0.1231 12.5 2000 0.2361 0.9293 0.9446 0.9369 0.9516
0.1231 14.06 2250 0.2194 0.9402 0.9528 0.9465 0.9576
0.0766 15.62 2500 0.2133 0.9416 0.9528 0.9472 0.9580
0.0766 17.19 2750 0.2117 0.9438 0.9558 0.9498 0.9597
0.0585 18.75 3000 0.2152 0.9417 0.9551 0.9483 0.9605
0.0585 20.31 3250 0.2070 0.9431 0.9551 0.9491 0.9588
0.0454 21.88 3500 0.2093 0.9489 0.9588 0.9538 0.9622
0.0454 23.44 3750 0.2034 0.9453 0.9566 0.9509 0.9610
0.0409 25.0 4000 0.2042 0.9445 0.9558 0.9501 0.9605

Framework versions

  • Transformers 4.21.2
  • Pytorch 1.12.1+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
3
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Evaluation results