<!-- 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. -->
EElayoutlmv3_jordyvl_rvl_cdip_100_examples_per_class_2023-09-22_went_gates
This model is a fine-tuned version of microsoft/layoutlmv3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9971
- Accuracy: 0.7275
- Exit 0 Accuracy: 0.0625
- Exit 1 Accuracy: 0.055
- Exit 2 Accuracy: 0.0525
- Exit 3 Accuracy: 0.0625
- Exit 4 Accuracy: 0.0625
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: 2e-05
- train_batch_size: 20
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 12
- total_train_batch_size: 240
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 60
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Exit 0 Accuracy | Exit 1 Accuracy | Exit 2 Accuracy | Exit 3 Accuracy | Exit 4 Accuracy |
---|---|---|---|---|---|---|---|---|---|
No log | 0.9 | 3 | 2.7374 | 0.1225 | 0.065 | 0.065 | 0.0825 | 0.0625 | 0.0625 |
No log | 1.9 | 6 | 2.6769 | 0.1525 | 0.0625 | 0.06 | 0.09 | 0.0625 | 0.0625 |
No log | 2.9 | 9 | 2.6398 | 0.1925 | 0.0625 | 0.0575 | 0.0875 | 0.0625 | 0.0625 |
No log | 3.9 | 12 | 2.6077 | 0.21 | 0.065 | 0.0525 | 0.085 | 0.0625 | 0.0625 |
No log | 4.9 | 15 | 2.5432 | 0.235 | 0.065 | 0.05 | 0.085 | 0.0625 | 0.0625 |
No log | 5.9 | 18 | 2.4882 | 0.2525 | 0.065 | 0.05 | 0.085 | 0.0625 | 0.0625 |
No log | 6.9 | 21 | 2.4322 | 0.27 | 0.065 | 0.04 | 0.085 | 0.0625 | 0.0625 |
No log | 7.9 | 24 | 2.3583 | 0.2925 | 0.065 | 0.0425 | 0.09 | 0.0625 | 0.0625 |
No log | 8.9 | 27 | 2.3032 | 0.3175 | 0.065 | 0.04 | 0.0925 | 0.0625 | 0.0625 |
No log | 9.9 | 30 | 2.2374 | 0.34 | 0.0675 | 0.04 | 0.0825 | 0.0625 | 0.0625 |
No log | 10.9 | 33 | 2.1630 | 0.3775 | 0.0675 | 0.0425 | 0.0775 | 0.0625 | 0.0625 |
No log | 11.9 | 36 | 2.0998 | 0.395 | 0.0675 | 0.045 | 0.07 | 0.0625 | 0.0625 |
No log | 12.9 | 39 | 2.0223 | 0.43 | 0.0675 | 0.045 | 0.0725 | 0.0625 | 0.0625 |
No log | 13.9 | 42 | 1.9143 | 0.4825 | 0.065 | 0.05 | 0.07 | 0.0625 | 0.0625 |
No log | 14.9 | 45 | 1.8329 | 0.5275 | 0.065 | 0.05 | 0.0725 | 0.0625 | 0.0625 |
No log | 15.9 | 48 | 1.7532 | 0.545 | 0.0625 | 0.05 | 0.065 | 0.0625 | 0.0625 |
No log | 16.9 | 51 | 1.6447 | 0.58 | 0.0625 | 0.05 | 0.065 | 0.0625 | 0.0625 |
No log | 17.9 | 54 | 1.5863 | 0.59 | 0.065 | 0.0475 | 0.065 | 0.0625 | 0.0625 |
No log | 18.9 | 57 | 1.5150 | 0.62 | 0.065 | 0.0475 | 0.0725 | 0.0625 | 0.0625 |
No log | 19.9 | 60 | 1.4389 | 0.6325 | 0.065 | 0.0475 | 0.0725 | 0.0625 | 0.0625 |
No log | 20.9 | 63 | 1.3834 | 0.6575 | 0.0625 | 0.0475 | 0.0725 | 0.0625 | 0.0625 |
No log | 21.9 | 66 | 1.3297 | 0.675 | 0.0625 | 0.0475 | 0.075 | 0.0625 | 0.0625 |
No log | 22.9 | 69 | 1.2805 | 0.6775 | 0.0625 | 0.0475 | 0.075 | 0.0625 | 0.0625 |
No log | 23.9 | 72 | 1.2448 | 0.6825 | 0.0625 | 0.05 | 0.07 | 0.0625 | 0.0625 |
No log | 24.9 | 75 | 1.1965 | 0.695 | 0.0625 | 0.05 | 0.0675 | 0.0625 | 0.0625 |
No log | 25.9 | 78 | 1.1793 | 0.6925 | 0.0625 | 0.0525 | 0.0675 | 0.0625 | 0.0625 |
No log | 26.9 | 81 | 1.1513 | 0.7 | 0.0625 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 27.9 | 84 | 1.1275 | 0.6975 | 0.0625 | 0.055 | 0.07 | 0.0625 | 0.0625 |
No log | 28.9 | 87 | 1.0880 | 0.7225 | 0.0625 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 29.9 | 90 | 1.0749 | 0.7175 | 0.065 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 30.9 | 93 | 1.0716 | 0.7175 | 0.0625 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 31.9 | 96 | 1.0496 | 0.72 | 0.0625 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 32.9 | 99 | 1.0253 | 0.7275 | 0.065 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 33.9 | 102 | 1.0305 | 0.7175 | 0.065 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 34.9 | 105 | 1.0273 | 0.73 | 0.065 | 0.055 | 0.0675 | 0.0625 | 0.0625 |
No log | 35.9 | 108 | 1.0072 | 0.73 | 0.065 | 0.055 | 0.07 | 0.0625 | 0.0625 |
No log | 36.9 | 111 | 1.0183 | 0.7225 | 0.0625 | 0.055 | 0.07 | 0.0625 | 0.0625 |
No log | 37.9 | 114 | 1.0079 | 0.725 | 0.0625 | 0.055 | 0.065 | 0.0625 | 0.0625 |
No log | 38.9 | 117 | 1.0086 | 0.7375 | 0.0625 | 0.055 | 0.065 | 0.0625 | 0.0625 |
No log | 39.9 | 120 | 0.9931 | 0.72 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 40.9 | 123 | 0.9968 | 0.7325 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 41.9 | 126 | 0.9848 | 0.73 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 42.9 | 129 | 0.9983 | 0.7325 | 0.06 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 43.9 | 132 | 0.9988 | 0.725 | 0.06 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 44.9 | 135 | 0.9941 | 0.7275 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 45.9 | 138 | 1.0053 | 0.72 | 0.0625 | 0.055 | 0.065 | 0.0625 | 0.0625 |
No log | 46.9 | 141 | 0.9904 | 0.725 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 47.9 | 144 | 0.9907 | 0.7225 | 0.0625 | 0.055 | 0.0625 | 0.0625 | 0.0625 |
No log | 48.9 | 147 | 0.9883 | 0.7325 | 0.06 | 0.055 | 0.06 | 0.0625 | 0.0625 |
No log | 49.9 | 150 | 0.9927 | 0.7325 | 0.0625 | 0.055 | 0.0575 | 0.0625 | 0.0625 |
No log | 50.9 | 153 | 0.9981 | 0.73 | 0.0625 | 0.055 | 0.055 | 0.0625 | 0.0625 |
No log | 51.9 | 156 | 0.9997 | 0.73 | 0.0625 | 0.055 | 0.055 | 0.0625 | 0.0625 |
No log | 52.9 | 159 | 0.9899 | 0.73 | 0.0575 | 0.055 | 0.055 | 0.0625 | 0.0625 |
No log | 53.9 | 162 | 0.9910 | 0.7275 | 0.0575 | 0.055 | 0.055 | 0.0625 | 0.0625 |
No log | 54.9 | 165 | 1.0002 | 0.725 | 0.0575 | 0.055 | 0.055 | 0.0625 | 0.0625 |
No log | 55.9 | 168 | 1.0044 | 0.7325 | 0.0575 | 0.055 | 0.0525 | 0.0625 | 0.0625 |
No log | 56.9 | 171 | 0.9977 | 0.73 | 0.0575 | 0.055 | 0.05 | 0.0625 | 0.0625 |
No log | 57.9 | 174 | 0.9960 | 0.73 | 0.0575 | 0.055 | 0.0525 | 0.0625 | 0.0625 |
No log | 58.9 | 177 | 0.9964 | 0.7275 | 0.06 | 0.055 | 0.0525 | 0.0625 | 0.0625 |
No log | 59.9 | 180 | 0.9971 | 0.7275 | 0.0625 | 0.055 | 0.0525 | 0.0625 | 0.0625 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2