<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
perioli_manifesti_v8.3
This model is a fine-tuned version of microsoft/layoutlmv3-base on the sroie dataset. It achieves the following results on the evaluation set:
- Loss: 0.0213
- Precision: 0.9600
- Recall: 0.9759
- F1: 0.9679
- Accuracy: 0.9962
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: 4000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.37 | 100 | 0.1334 | 0.7328 | 0.7546 | 0.7435 | 0.9605 |
No log | 0.74 | 200 | 0.0562 | 0.8838 | 0.9128 | 0.8981 | 0.9862 |
No log | 1.11 | 300 | 0.0520 | 0.8679 | 0.9311 | 0.8984 | 0.9858 |
No log | 1.48 | 400 | 0.0392 | 0.8995 | 0.9345 | 0.9167 | 0.9888 |
0.149 | 1.85 | 500 | 0.0306 | 0.9235 | 0.9535 | 0.9383 | 0.9921 |
0.149 | 2.22 | 600 | 0.0263 | 0.9399 | 0.9542 | 0.9470 | 0.9933 |
0.149 | 2.59 | 700 | 0.0253 | 0.9404 | 0.9576 | 0.9489 | 0.9935 |
0.149 | 2.96 | 800 | 0.0243 | 0.9407 | 0.9576 | 0.9491 | 0.9935 |
0.149 | 3.33 | 900 | 0.0276 | 0.9225 | 0.9607 | 0.9412 | 0.9927 |
0.0267 | 3.7 | 1000 | 0.0252 | 0.9329 | 0.9638 | 0.9481 | 0.9936 |
0.0267 | 4.07 | 1100 | 0.0186 | 0.9517 | 0.9704 | 0.9609 | 0.9953 |
0.0267 | 4.44 | 1200 | 0.0218 | 0.9506 | 0.9624 | 0.9565 | 0.9946 |
0.0267 | 4.81 | 1300 | 0.0188 | 0.9567 | 0.9666 | 0.9616 | 0.9954 |
0.0267 | 5.19 | 1400 | 0.0195 | 0.9573 | 0.9728 | 0.9650 | 0.9957 |
0.0153 | 5.56 | 1500 | 0.0238 | 0.9526 | 0.9628 | 0.9577 | 0.9946 |
0.0153 | 5.93 | 1600 | 0.0211 | 0.9469 | 0.9704 | 0.9585 | 0.9950 |
0.0153 | 6.3 | 1700 | 0.0218 | 0.9474 | 0.9679 | 0.9575 | 0.9949 |
0.0153 | 6.67 | 1800 | 0.0218 | 0.9478 | 0.9700 | 0.9588 | 0.9951 |
0.0153 | 7.04 | 1900 | 0.0204 | 0.9559 | 0.9707 | 0.9632 | 0.9955 |
0.01 | 7.41 | 2000 | 0.0211 | 0.9498 | 0.9728 | 0.9612 | 0.9954 |
0.01 | 7.78 | 2100 | 0.0198 | 0.9589 | 0.9741 | 0.9665 | 0.9960 |
0.01 | 8.15 | 2200 | 0.0187 | 0.9609 | 0.9738 | 0.9673 | 0.9961 |
0.01 | 8.52 | 2300 | 0.0193 | 0.9606 | 0.9759 | 0.9682 | 0.9963 |
0.01 | 8.89 | 2400 | 0.0212 | 0.9540 | 0.9717 | 0.9628 | 0.9956 |
0.0071 | 9.26 | 2500 | 0.0203 | 0.9579 | 0.9728 | 0.9653 | 0.9959 |
0.0071 | 9.63 | 2600 | 0.0200 | 0.9590 | 0.9752 | 0.9670 | 0.9960 |
0.0071 | 10.0 | 2700 | 0.0215 | 0.9515 | 0.9735 | 0.9623 | 0.9955 |
0.0071 | 10.37 | 2800 | 0.0209 | 0.9586 | 0.9748 | 0.9667 | 0.9960 |
0.0071 | 10.74 | 2900 | 0.0191 | 0.9589 | 0.9741 | 0.9665 | 0.9961 |
0.0051 | 11.11 | 3000 | 0.0194 | 0.9606 | 0.9738 | 0.9671 | 0.9961 |
0.0051 | 11.48 | 3100 | 0.0209 | 0.9561 | 0.9755 | 0.9657 | 0.9961 |
0.0051 | 11.85 | 3200 | 0.0202 | 0.9580 | 0.9741 | 0.9660 | 0.9961 |
0.0051 | 12.22 | 3300 | 0.0210 | 0.9589 | 0.9738 | 0.9663 | 0.9961 |
0.0051 | 12.59 | 3400 | 0.0200 | 0.9583 | 0.9745 | 0.9663 | 0.9960 |
0.0041 | 12.96 | 3500 | 0.0209 | 0.9577 | 0.9748 | 0.9662 | 0.9960 |
0.0041 | 13.33 | 3600 | 0.0209 | 0.9603 | 0.9752 | 0.9677 | 0.9962 |
0.0041 | 13.7 | 3700 | 0.0215 | 0.9583 | 0.9752 | 0.9667 | 0.9961 |
0.0041 | 14.07 | 3800 | 0.0215 | 0.9577 | 0.9748 | 0.9662 | 0.9960 |
0.0041 | 14.44 | 3900 | 0.0213 | 0.9600 | 0.9759 | 0.9679 | 0.9962 |
0.003 | 14.81 | 4000 | 0.0213 | 0.9600 | 0.9759 | 0.9679 | 0.9962 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3