<!-- 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. -->
Bol-4.0-invoicefromclients_LOC_CAD
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: 0.7295
- Precision: 0.5245
- Recall: 0.4697
- F1: 0.4956
- Accuracy: 0.8690
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: 1500
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 0.32 | 100 | 1.3578 | 0.3480 | 0.0304 | 0.0560 | 0.7636 |
| No log | 0.63 | 200 | 1.0269 | 0.1777 | 0.0748 | 0.1052 | 0.7962 |
| No log | 0.95 | 300 | 0.8968 | 0.3288 | 0.1869 | 0.2383 | 0.8180 |
| No log | 1.27 | 400 | 0.8574 | 0.3945 | 0.2212 | 0.2835 | 0.8227 |
| 0.8908 | 1.58 | 500 | 0.7533 | 0.3144 | 0.2709 | 0.2910 | 0.8181 |
| 0.8908 | 1.9 | 600 | 0.7001 | 0.3913 | 0.3106 | 0.3463 | 0.8414 |
| 0.8908 | 2.22 | 700 | 0.6915 | 0.4998 | 0.3869 | 0.4361 | 0.8572 |
| 0.8908 | 2.53 | 800 | 0.7375 | 0.4331 | 0.3703 | 0.3993 | 0.8475 |
| 0.8908 | 2.85 | 900 | 0.6590 | 0.4682 | 0.3973 | 0.4299 | 0.8633 |
| 0.353 | 3.16 | 1000 | 0.7389 | 0.5479 | 0.4274 | 0.4802 | 0.8650 |
| 0.353 | 3.48 | 1100 | 0.7387 | 0.5568 | 0.4474 | 0.4962 | 0.8635 |
| 0.353 | 3.8 | 1200 | 0.6881 | 0.5011 | 0.4539 | 0.4763 | 0.8707 |
| 0.353 | 4.11 | 1300 | 0.6881 | 0.5159 | 0.4624 | 0.4877 | 0.8684 |
| 0.353 | 4.43 | 1400 | 0.7308 | 0.5532 | 0.4751 | 0.5112 | 0.8713 |
| 0.1947 | 4.75 | 1500 | 0.7295 | 0.5245 | 0.4697 | 0.4956 | 0.8690 |
Framework versions
- Transformers 4.28.0.dev0
- Pytorch 2.0.0+cu118
- Datasets 2.2.2
- Tokenizers 0.13.2
