<!-- 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_vgm_v5.2
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: 0.0093
- Precision: 0.9702
- Recall: 0.9540
- F1: 0.9620
- Accuracy: 0.9980
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: 2000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.53 | 100 | 0.0787 | 0.5595 | 0.3933 | 0.4619 | 0.9811 |
No log | 1.06 | 200 | 0.0450 | 0.8069 | 0.6820 | 0.7392 | 0.9909 |
No log | 1.6 | 300 | 0.0438 | 0.6469 | 0.8201 | 0.7232 | 0.9874 |
No log | 2.13 | 400 | 0.0230 | 0.8063 | 0.8536 | 0.8293 | 0.9935 |
0.0742 | 2.66 | 500 | 0.0177 | 0.8214 | 0.8661 | 0.8432 | 0.9946 |
0.0742 | 3.19 | 600 | 0.0251 | 0.8295 | 0.8954 | 0.8612 | 0.9943 |
0.0742 | 3.72 | 700 | 0.0167 | 0.8782 | 0.8745 | 0.8763 | 0.9961 |
0.0742 | 4.26 | 800 | 0.0183 | 0.8510 | 0.9079 | 0.8785 | 0.9949 |
0.0742 | 4.79 | 900 | 0.0100 | 0.9481 | 0.9163 | 0.9319 | 0.9969 |
0.0085 | 5.32 | 1000 | 0.0153 | 0.9048 | 0.9540 | 0.9287 | 0.9967 |
0.0085 | 5.85 | 1100 | 0.0112 | 0.9253 | 0.9331 | 0.9292 | 0.9972 |
0.0085 | 6.38 | 1200 | 0.0120 | 0.9184 | 0.9414 | 0.9298 | 0.9971 |
0.0085 | 6.91 | 1300 | 0.0072 | 0.9417 | 0.9456 | 0.9436 | 0.9980 |
0.0085 | 7.45 | 1400 | 0.0102 | 0.9617 | 0.9456 | 0.9536 | 0.9977 |
0.0029 | 7.98 | 1500 | 0.0101 | 0.9702 | 0.9540 | 0.9620 | 0.9980 |
0.0029 | 8.51 | 1600 | 0.0098 | 0.9580 | 0.9540 | 0.9560 | 0.9979 |
0.0029 | 9.04 | 1700 | 0.0075 | 0.9536 | 0.9456 | 0.9496 | 0.9980 |
0.0029 | 9.57 | 1800 | 0.0083 | 0.9702 | 0.9540 | 0.9620 | 0.9980 |
0.0029 | 10.11 | 1900 | 0.0093 | 0.9617 | 0.9456 | 0.9536 | 0.9979 |
0.001 | 10.64 | 2000 | 0.0093 | 0.9702 | 0.9540 | 0.9620 | 0.9980 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3