<!-- 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_v8.4
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: 0.0189
- Precision: 0.9607
- Recall: 0.9772
- F1: 0.9689
- Accuracy: 0.9963
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 | 0.33 | 100 | 0.1293 | 0.7382 | 0.7649 | 0.7513 | 0.9604 |
No log | 0.65 | 200 | 0.0680 | 0.8235 | 0.8845 | 0.8529 | 0.9783 |
No log | 0.98 | 300 | 0.0408 | 0.8919 | 0.9297 | 0.9104 | 0.9881 |
No log | 1.31 | 400 | 0.0335 | 0.9051 | 0.9500 | 0.9270 | 0.9905 |
0.1635 | 1.63 | 500 | 0.0372 | 0.8970 | 0.9431 | 0.9195 | 0.9895 |
0.1635 | 1.96 | 600 | 0.0251 | 0.9333 | 0.9600 | 0.9465 | 0.9932 |
0.1635 | 2.29 | 700 | 0.0284 | 0.9277 | 0.9552 | 0.9412 | 0.9923 |
0.1635 | 2.61 | 800 | 0.0237 | 0.9317 | 0.9645 | 0.9478 | 0.9935 |
0.1635 | 2.94 | 900 | 0.0210 | 0.9398 | 0.9624 | 0.9510 | 0.9939 |
0.0277 | 3.27 | 1000 | 0.0200 | 0.9503 | 0.9683 | 0.9592 | 0.9949 |
0.0277 | 3.59 | 1100 | 0.0201 | 0.9470 | 0.9676 | 0.9572 | 0.9950 |
0.0277 | 3.92 | 1200 | 0.0197 | 0.9567 | 0.9669 | 0.9618 | 0.9954 |
0.0277 | 4.25 | 1300 | 0.0209 | 0.9472 | 0.9717 | 0.9593 | 0.9952 |
0.0277 | 4.58 | 1400 | 0.0212 | 0.9561 | 0.9748 | 0.9654 | 0.9956 |
0.0153 | 4.9 | 1500 | 0.0227 | 0.9469 | 0.9710 | 0.9588 | 0.9949 |
0.0153 | 5.23 | 1600 | 0.0238 | 0.9434 | 0.9717 | 0.9574 | 0.9950 |
0.0153 | 5.56 | 1700 | 0.0228 | 0.9476 | 0.9721 | 0.9597 | 0.9951 |
0.0153 | 5.88 | 1800 | 0.0206 | 0.9541 | 0.9738 | 0.9638 | 0.9956 |
0.0153 | 6.21 | 1900 | 0.0192 | 0.9579 | 0.9714 | 0.9646 | 0.9956 |
0.0109 | 6.54 | 2000 | 0.0198 | 0.9593 | 0.9745 | 0.9668 | 0.9960 |
0.0109 | 6.86 | 2100 | 0.0207 | 0.9557 | 0.9731 | 0.9643 | 0.9957 |
0.0109 | 7.19 | 2200 | 0.0215 | 0.9614 | 0.9700 | 0.9657 | 0.9959 |
0.0109 | 7.52 | 2300 | 0.0226 | 0.9528 | 0.9752 | 0.9639 | 0.9959 |
0.0109 | 7.84 | 2400 | 0.0218 | 0.9563 | 0.9738 | 0.9650 | 0.9959 |
0.0077 | 8.17 | 2500 | 0.0209 | 0.9587 | 0.9766 | 0.9676 | 0.9963 |
0.0077 | 8.5 | 2600 | 0.0236 | 0.9469 | 0.9710 | 0.9588 | 0.9951 |
0.0077 | 8.82 | 2700 | 0.0194 | 0.9570 | 0.9748 | 0.9658 | 0.9960 |
0.0077 | 9.15 | 2800 | 0.0196 | 0.9528 | 0.9745 | 0.9635 | 0.9958 |
0.0077 | 9.48 | 2900 | 0.0199 | 0.9590 | 0.9748 | 0.9668 | 0.9961 |
0.0057 | 9.8 | 3000 | 0.0188 | 0.9613 | 0.9766 | 0.9689 | 0.9963 |
0.0057 | 10.13 | 3100 | 0.0201 | 0.9551 | 0.9759 | 0.9654 | 0.9959 |
0.0057 | 10.46 | 3200 | 0.0202 | 0.9544 | 0.9731 | 0.9636 | 0.9957 |
0.0057 | 10.78 | 3300 | 0.0202 | 0.9555 | 0.9759 | 0.9656 | 0.9959 |
0.0057 | 11.11 | 3400 | 0.0173 | 0.9610 | 0.9766 | 0.9687 | 0.9963 |
0.0045 | 11.44 | 3500 | 0.0191 | 0.9558 | 0.9755 | 0.9655 | 0.9960 |
0.0045 | 11.76 | 3600 | 0.0188 | 0.9557 | 0.9752 | 0.9654 | 0.9959 |
0.0045 | 12.09 | 3700 | 0.0187 | 0.9567 | 0.9755 | 0.9660 | 0.9960 |
0.0045 | 12.42 | 3800 | 0.0190 | 0.9558 | 0.9755 | 0.9655 | 0.9960 |
0.0045 | 12.75 | 3900 | 0.0192 | 0.9577 | 0.9762 | 0.9669 | 0.9961 |
0.004 | 13.07 | 4000 | 0.0189 | 0.9607 | 0.9772 | 0.9689 | 0.9963 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3