<!-- 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_v5.1
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: 0.0301
- Precision: 0.9215
- Recall: 0.9497
- F1: 0.9354
- Accuracy: 0.9922
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: 2700
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.48 | 100 | 0.1283 | 0.7806 | 0.8086 | 0.7944 | 0.9734 |
No log | 0.96 | 200 | 0.0539 | 0.8832 | 0.9102 | 0.8965 | 0.9865 |
No log | 1.44 | 300 | 0.0451 | 0.9001 | 0.9389 | 0.9191 | 0.9893 |
No log | 1.92 | 400 | 0.0395 | 0.9058 | 0.9326 | 0.9190 | 0.9890 |
0.1374 | 2.4 | 500 | 0.0312 | 0.9269 | 0.9461 | 0.9364 | 0.9926 |
0.1374 | 2.88 | 600 | 0.0306 | 0.9230 | 0.9587 | 0.9405 | 0.9925 |
0.1374 | 3.37 | 700 | 0.0260 | 0.9356 | 0.9659 | 0.9505 | 0.9939 |
0.1374 | 3.85 | 800 | 0.0331 | 0.9244 | 0.9668 | 0.9451 | 0.9928 |
0.1374 | 4.33 | 900 | 0.0296 | 0.9099 | 0.9524 | 0.9306 | 0.9909 |
0.0233 | 4.81 | 1000 | 0.0235 | 0.9379 | 0.9632 | 0.9504 | 0.9943 |
0.0233 | 5.29 | 1100 | 0.0200 | 0.9407 | 0.9695 | 0.9549 | 0.9948 |
0.0233 | 5.77 | 1200 | 0.0263 | 0.9233 | 0.9632 | 0.9428 | 0.9929 |
0.0233 | 6.25 | 1300 | 0.0267 | 0.9291 | 0.9650 | 0.9467 | 0.9933 |
0.0233 | 6.73 | 1400 | 0.0243 | 0.9348 | 0.9668 | 0.9505 | 0.9941 |
0.0147 | 7.21 | 1500 | 0.0260 | 0.9280 | 0.9614 | 0.9444 | 0.9934 |
0.0147 | 7.69 | 1600 | 0.0256 | 0.9311 | 0.9596 | 0.9451 | 0.9934 |
0.0147 | 8.17 | 1700 | 0.0246 | 0.9291 | 0.9542 | 0.9415 | 0.9930 |
0.0147 | 8.65 | 1800 | 0.0242 | 0.9364 | 0.9650 | 0.9504 | 0.9941 |
0.0147 | 9.13 | 1900 | 0.0266 | 0.9268 | 0.9560 | 0.9412 | 0.9928 |
0.0107 | 9.62 | 2000 | 0.0300 | 0.9296 | 0.9605 | 0.9448 | 0.9932 |
0.0107 | 10.1 | 2100 | 0.0277 | 0.9365 | 0.9677 | 0.9518 | 0.9941 |
0.0107 | 10.58 | 2200 | 0.0292 | 0.9324 | 0.9668 | 0.9493 | 0.9937 |
0.0107 | 11.06 | 2300 | 0.0247 | 0.9284 | 0.9560 | 0.9420 | 0.9930 |
0.0107 | 11.54 | 2400 | 0.0298 | 0.9191 | 0.9488 | 0.9337 | 0.9921 |
0.0067 | 12.02 | 2500 | 0.0303 | 0.9215 | 0.9497 | 0.9354 | 0.9922 |
0.0067 | 12.5 | 2600 | 0.0305 | 0.9234 | 0.9533 | 0.9381 | 0.9925 |
0.0067 | 12.98 | 2700 | 0.0301 | 0.9215 | 0.9497 | 0.9354 | 0.9922 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3