Edit model card

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: 0.2213
  • Precision: 0.9329
  • Recall: 0.9469
  • F1: 0.9398
  • Accuracy: 0.9516

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: 2500

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.56 250 1.0664 0.6765 0.7530 0.7127 0.7818
1.4379 3.12 500 0.6115 0.8199 0.8518 0.8355 0.8646
1.4379 4.69 750 0.4192 0.8794 0.9004 0.8898 0.9028
0.4232 6.25 1000 0.3239 0.9180 0.9296 0.9238 0.9304
0.4232 7.81 1250 0.2840 0.9197 0.9341 0.9268 0.9389
0.2273 9.38 1500 0.2562 0.9217 0.9341 0.9279 0.9376
0.2273 10.94 1750 0.2574 0.9304 0.9401 0.9352 0.9410
0.157 12.5 2000 0.2327 0.9293 0.9439 0.9365 0.9482
0.157 14.06 2250 0.2217 0.9351 0.9491 0.9421 0.9520
0.1208 15.62 2500 0.2213 0.9329 0.9469 0.9398 0.9516

Framework versions

  • Transformers 4.23.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.6.1
  • Tokenizers 0.13.1
Downloads last month
6
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