<!-- 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_2023-07-07_went-g040
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: 1.1113
- Accuracy: 0.71
- Exit 0 Accuracy: 0.115
- Exit 1 Accuracy: 0.15
- Exit 2 Accuracy: 0.2025
- 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: 12
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 24
- total_train_batch_size: 288
- 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 | 0.72 | 2 | 2.7602 | 0.11 | 0.105 | 0.0675 | 0.0825 | 0.0625 | 0.0625 |
No log | 1.72 | 4 | 2.7300 | 0.115 | 0.105 | 0.065 | 0.09 | 0.0625 | 0.0625 |
No log | 2.72 | 6 | 2.6942 | 0.135 | 0.1075 | 0.06 | 0.105 | 0.0625 | 0.0625 |
No log | 3.72 | 8 | 2.6645 | 0.1725 | 0.145 | 0.055 | 0.1125 | 0.0625 | 0.0625 |
No log | 4.72 | 10 | 2.6341 | 0.175 | 0.1275 | 0.06 | 0.1175 | 0.0625 | 0.0625 |
No log | 5.72 | 12 | 2.5949 | 0.215 | 0.125 | 0.08 | 0.1125 | 0.0625 | 0.0625 |
No log | 6.72 | 14 | 2.5729 | 0.2025 | 0.12 | 0.08 | 0.1175 | 0.0625 | 0.0625 |
No log | 7.72 | 16 | 2.5500 | 0.2075 | 0.115 | 0.09 | 0.125 | 0.0625 | 0.0625 |
No log | 8.72 | 18 | 2.5220 | 0.2175 | 0.1175 | 0.0925 | 0.125 | 0.0625 | 0.0625 |
No log | 9.72 | 20 | 2.4976 | 0.2275 | 0.1225 | 0.0975 | 0.125 | 0.0625 | 0.0625 |
No log | 10.72 | 22 | 2.4523 | 0.2525 | 0.1225 | 0.1 | 0.125 | 0.0625 | 0.0625 |
No log | 11.72 | 24 | 2.3993 | 0.295 | 0.12 | 0.1275 | 0.1225 | 0.0625 | 0.0625 |
No log | 12.72 | 26 | 2.3545 | 0.315 | 0.12 | 0.1175 | 0.125 | 0.0625 | 0.0625 |
No log | 13.72 | 28 | 2.3057 | 0.335 | 0.1175 | 0.1175 | 0.1225 | 0.0625 | 0.0625 |
No log | 14.72 | 30 | 2.2490 | 0.355 | 0.1175 | 0.1275 | 0.1275 | 0.0625 | 0.0625 |
No log | 15.72 | 32 | 2.2131 | 0.355 | 0.115 | 0.125 | 0.1225 | 0.0625 | 0.0625 |
No log | 16.72 | 34 | 2.1526 | 0.3725 | 0.1125 | 0.135 | 0.125 | 0.0625 | 0.0625 |
No log | 17.72 | 36 | 2.0828 | 0.3975 | 0.1025 | 0.14 | 0.125 | 0.0625 | 0.0625 |
No log | 18.72 | 38 | 2.0196 | 0.4225 | 0.1075 | 0.1425 | 0.1275 | 0.0625 | 0.0625 |
No log | 19.72 | 40 | 1.9756 | 0.4275 | 0.11 | 0.14 | 0.13 | 0.0625 | 0.0625 |
No log | 20.72 | 42 | 1.9239 | 0.4625 | 0.1125 | 0.1425 | 0.1275 | 0.0625 | 0.0625 |
No log | 21.72 | 44 | 1.8449 | 0.505 | 0.11 | 0.14 | 0.1275 | 0.0625 | 0.0625 |
No log | 22.72 | 46 | 1.7852 | 0.53 | 0.11 | 0.14 | 0.1275 | 0.0625 | 0.0625 |
No log | 23.72 | 48 | 1.7626 | 0.5325 | 0.11 | 0.1425 | 0.1375 | 0.0625 | 0.0625 |
No log | 24.72 | 50 | 1.7041 | 0.5575 | 0.11 | 0.145 | 0.1475 | 0.0625 | 0.0625 |
No log | 25.72 | 52 | 1.6443 | 0.5825 | 0.1075 | 0.1475 | 0.145 | 0.0625 | 0.0625 |
No log | 26.72 | 54 | 1.6042 | 0.6 | 0.1075 | 0.1475 | 0.145 | 0.0625 | 0.0625 |
No log | 27.72 | 56 | 1.5753 | 0.6 | 0.1075 | 0.1475 | 0.15 | 0.0625 | 0.0625 |
No log | 28.72 | 58 | 1.5241 | 0.615 | 0.1075 | 0.145 | 0.1525 | 0.0625 | 0.0625 |
No log | 29.72 | 60 | 1.4874 | 0.6225 | 0.115 | 0.1425 | 0.155 | 0.0625 | 0.0625 |
No log | 30.72 | 62 | 1.4638 | 0.6275 | 0.115 | 0.145 | 0.1525 | 0.0625 | 0.0625 |
No log | 31.72 | 64 | 1.4460 | 0.64 | 0.1125 | 0.145 | 0.1525 | 0.0625 | 0.0625 |
No log | 32.72 | 66 | 1.3980 | 0.655 | 0.1125 | 0.145 | 0.1525 | 0.0625 | 0.0625 |
No log | 33.72 | 68 | 1.3708 | 0.6425 | 0.11 | 0.145 | 0.155 | 0.0625 | 0.0625 |
No log | 34.72 | 70 | 1.3584 | 0.6575 | 0.11 | 0.145 | 0.1575 | 0.0625 | 0.0625 |
No log | 35.72 | 72 | 1.3339 | 0.66 | 0.1125 | 0.1475 | 0.16 | 0.0625 | 0.0625 |
No log | 36.72 | 74 | 1.3046 | 0.6725 | 0.1125 | 0.15 | 0.1625 | 0.0625 | 0.0625 |
No log | 37.72 | 76 | 1.2891 | 0.6675 | 0.115 | 0.15 | 0.1625 | 0.0625 | 0.0625 |
No log | 38.72 | 78 | 1.2684 | 0.68 | 0.115 | 0.15 | 0.165 | 0.0625 | 0.0625 |
No log | 39.72 | 80 | 1.2400 | 0.705 | 0.1175 | 0.15 | 0.175 | 0.0625 | 0.0625 |
No log | 40.72 | 82 | 1.2277 | 0.695 | 0.12 | 0.15 | 0.175 | 0.0625 | 0.0625 |
No log | 41.72 | 84 | 1.2234 | 0.6975 | 0.1175 | 0.15 | 0.175 | 0.0625 | 0.0625 |
No log | 42.72 | 86 | 1.2082 | 0.6925 | 0.115 | 0.15 | 0.175 | 0.0625 | 0.0625 |
No log | 43.72 | 88 | 1.1851 | 0.71 | 0.1175 | 0.15 | 0.1725 | 0.0625 | 0.0625 |
No log | 44.72 | 90 | 1.1743 | 0.7075 | 0.1175 | 0.15 | 0.1725 | 0.0625 | 0.0625 |
No log | 45.72 | 92 | 1.1764 | 0.7 | 0.1175 | 0.15 | 0.1725 | 0.0625 | 0.0625 |
No log | 46.72 | 94 | 1.1731 | 0.6975 | 0.1175 | 0.1525 | 0.1775 | 0.0625 | 0.0625 |
No log | 47.72 | 96 | 1.1512 | 0.6975 | 0.1175 | 0.1525 | 0.175 | 0.0625 | 0.0625 |
No log | 48.72 | 98 | 1.1382 | 0.705 | 0.1175 | 0.1525 | 0.1775 | 0.0625 | 0.0625 |
No log | 49.72 | 100 | 1.1405 | 0.7 | 0.115 | 0.1525 | 0.1775 | 0.0625 | 0.0625 |
No log | 50.72 | 102 | 1.1434 | 0.71 | 0.115 | 0.1525 | 0.1875 | 0.0625 | 0.0625 |
No log | 51.72 | 104 | 1.1324 | 0.71 | 0.115 | 0.1525 | 0.19 | 0.0625 | 0.0625 |
No log | 52.72 | 106 | 1.1216 | 0.7125 | 0.115 | 0.1525 | 0.195 | 0.0625 | 0.0625 |
No log | 53.72 | 108 | 1.1166 | 0.7075 | 0.115 | 0.1525 | 0.2 | 0.0625 | 0.0625 |
No log | 54.72 | 110 | 1.1134 | 0.705 | 0.1125 | 0.1525 | 0.1975 | 0.0625 | 0.0625 |
No log | 55.72 | 112 | 1.1127 | 0.7025 | 0.1125 | 0.1525 | 0.2 | 0.0625 | 0.0625 |
No log | 56.72 | 114 | 1.1133 | 0.705 | 0.1125 | 0.1525 | 0.2025 | 0.0625 | 0.0625 |
No log | 57.72 | 116 | 1.1127 | 0.705 | 0.1125 | 0.15 | 0.2025 | 0.0625 | 0.0625 |
No log | 58.72 | 118 | 1.1116 | 0.7075 | 0.115 | 0.15 | 0.2025 | 0.0625 | 0.0625 |
No log | 59.72 | 120 | 1.1113 | 0.71 | 0.115 | 0.15 | 0.2025 | 0.0625 | 0.0625 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2