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: 0.2268
- Precision: 0.9386
- Recall: 0.9491
- F1: 0.9438
- Accuracy: 0.9542
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 | 1.0455 | 0.7145 | 0.7867 | 0.7488 | 0.7886 |
1.4126 | 3.12 | 500 | 0.5824 | 0.8322 | 0.8615 | 0.8466 | 0.8697 |
1.4126 | 4.69 | 750 | 0.4120 | 0.8620 | 0.8975 | 0.8794 | 0.9053 |
0.4193 | 6.25 | 1000 | 0.3474 | 0.9020 | 0.9229 | 0.9123 | 0.9283 |
0.4193 | 7.81 | 1250 | 0.2864 | 0.9210 | 0.9341 | 0.9275 | 0.9406 |
0.2287 | 9.38 | 1500 | 0.2501 | 0.9290 | 0.9401 | 0.9345 | 0.9457 |
0.2287 | 10.94 | 1750 | 0.2451 | 0.9270 | 0.9409 | 0.9339 | 0.9469 |
0.1607 | 12.5 | 2000 | 0.2408 | 0.9348 | 0.9446 | 0.9397 | 0.9520 |
0.1607 | 14.06 | 2250 | 0.2296 | 0.9328 | 0.9461 | 0.9394 | 0.9537 |
0.1235 | 15.62 | 2500 | 0.2268 | 0.9386 | 0.9491 | 0.9438 | 0.9542 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
- Downloads last month
- 8
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Evaluation results
- Precision on cord-layoutlmv3test set self-reported0.939
- Recall on cord-layoutlmv3test set self-reported0.949
- F1 on cord-layoutlmv3test set self-reported0.944
- Accuracy on cord-layoutlmv3test set self-reported0.954