<!-- 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_2023-08-27_txt_vis_conc_enc_9_10_11_12_ramp
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: 1.3922
- Accuracy: 0.7675
- Exit 0 Accuracy: 0.08
- Exit 1 Accuracy: 0.7675
- Exit 2 Accuracy: 0.7675
- Exit 3 Accuracy: 0.7675
- Exit 4 Accuracy: 0.765
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: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 24
- total_train_batch_size: 48
- 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 | 0.96 | 16 | 2.6822 | 0.1625 | 0.07 | 0.0625 | 0.0625 | 0.0625 | 0.1 |
No log | 1.98 | 33 | 2.5295 | 0.235 | 0.0775 | 0.0625 | 0.0625 | 0.0625 | 0.1625 |
No log | 3.0 | 50 | 2.3203 | 0.3175 | 0.0825 | 0.0625 | 0.0625 | 0.0625 | 0.2575 |
No log | 3.96 | 66 | 2.0766 | 0.4075 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.3375 |
No log | 4.98 | 83 | 1.7926 | 0.57 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.4675 |
No log | 6.0 | 100 | 1.5171 | 0.6225 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.4925 |
No log | 6.96 | 116 | 1.3587 | 0.665 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.575 |
No log | 7.98 | 133 | 1.1768 | 0.7275 | 0.0825 | 0.0625 | 0.0625 | 0.0625 | 0.6175 |
No log | 9.0 | 150 | 1.0821 | 0.73 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.665 |
No log | 9.96 | 166 | 1.0058 | 0.745 | 0.085 | 0.0625 | 0.0625 | 0.0625 | 0.7125 |
No log | 10.98 | 183 | 0.9673 | 0.7475 | 0.0875 | 0.0625 | 0.0625 | 0.0625 | 0.7275 |
No log | 12.0 | 200 | 0.9760 | 0.7475 | 0.085 | 0.0625 | 0.0625 | 0.07 | 0.74 |
No log | 12.96 | 216 | 0.9375 | 0.7575 | 0.085 | 0.0625 | 0.0625 | 0.0775 | 0.745 |
No log | 13.98 | 233 | 0.9334 | 0.7525 | 0.0825 | 0.0625 | 0.0625 | 0.1125 | 0.7475 |
No log | 15.0 | 250 | 0.9384 | 0.7675 | 0.0725 | 0.0625 | 0.0625 | 0.12 | 0.76 |
No log | 15.96 | 266 | 0.9548 | 0.7625 | 0.0725 | 0.0625 | 0.0625 | 0.1225 | 0.7775 |
No log | 16.98 | 283 | 1.0016 | 0.7675 | 0.0625 | 0.0625 | 0.0625 | 0.2875 | 0.7575 |
No log | 18.0 | 300 | 1.0385 | 0.765 | 0.065 | 0.0625 | 0.0625 | 0.4675 | 0.7625 |
No log | 18.96 | 316 | 1.1031 | 0.755 | 0.065 | 0.0625 | 0.1225 | 0.68 | 0.76 |
No log | 19.98 | 333 | 1.0838 | 0.77 | 0.065 | 0.0625 | 0.175 | 0.7625 | 0.765 |
No log | 21.0 | 350 | 1.1351 | 0.76 | 0.0675 | 0.0625 | 0.6875 | 0.765 | 0.7575 |
No log | 21.96 | 366 | 1.1676 | 0.75 | 0.0675 | 0.1325 | 0.735 | 0.7525 | 0.75 |
No log | 22.98 | 383 | 1.1791 | 0.7675 | 0.06 | 0.53 | 0.7475 | 0.7725 | 0.7725 |
No log | 24.0 | 400 | 1.1515 | 0.77 | 0.065 | 0.755 | 0.7625 | 0.77 | 0.765 |
No log | 24.96 | 416 | 1.1810 | 0.7625 | 0.065 | 0.7575 | 0.7675 | 0.77 | 0.7675 |
No log | 25.98 | 433 | 1.2599 | 0.7625 | 0.065 | 0.76 | 0.7575 | 0.755 | 0.77 |
No log | 27.0 | 450 | 1.2142 | 0.7775 | 0.07 | 0.76 | 0.765 | 0.775 | 0.7775 |
No log | 27.96 | 466 | 1.2431 | 0.77 | 0.0725 | 0.7525 | 0.765 | 0.7725 | 0.7725 |
No log | 28.98 | 483 | 1.2692 | 0.7625 | 0.0725 | 0.76 | 0.7625 | 0.7675 | 0.76 |
1.3506 | 30.0 | 500 | 1.2868 | 0.7625 | 0.0725 | 0.765 | 0.765 | 0.77 | 0.765 |
1.3506 | 30.96 | 516 | 1.3143 | 0.76 | 0.0725 | 0.7625 | 0.7625 | 0.7675 | 0.76 |
1.3506 | 31.98 | 533 | 1.3064 | 0.77 | 0.0725 | 0.7675 | 0.77 | 0.7725 | 0.765 |
1.3506 | 33.0 | 550 | 1.3295 | 0.765 | 0.0725 | 0.76 | 0.7675 | 0.76 | 0.7625 |
1.3506 | 33.96 | 566 | 1.3529 | 0.76 | 0.0675 | 0.7675 | 0.755 | 0.7675 | 0.76 |
1.3506 | 34.98 | 583 | 1.3487 | 0.7675 | 0.065 | 0.76 | 0.7625 | 0.7625 | 0.7625 |
1.3506 | 36.0 | 600 | 1.3456 | 0.7725 | 0.065 | 0.76 | 0.765 | 0.7675 | 0.7675 |
1.3506 | 36.96 | 616 | 1.3568 | 0.7675 | 0.0725 | 0.7625 | 0.7675 | 0.765 | 0.765 |
1.3506 | 37.98 | 633 | 1.3626 | 0.765 | 0.075 | 0.76 | 0.77 | 0.77 | 0.765 |
1.3506 | 39.0 | 650 | 1.3631 | 0.7675 | 0.075 | 0.76 | 0.7675 | 0.77 | 0.7625 |
1.3506 | 39.96 | 666 | 1.3707 | 0.765 | 0.075 | 0.765 | 0.7675 | 0.7725 | 0.765 |
1.3506 | 40.98 | 683 | 1.3761 | 0.765 | 0.0775 | 0.76 | 0.7675 | 0.77 | 0.765 |
1.3506 | 42.0 | 700 | 1.3745 | 0.77 | 0.08 | 0.76 | 0.765 | 0.77 | 0.765 |
1.3506 | 42.96 | 716 | 1.3685 | 0.77 | 0.08 | 0.7625 | 0.765 | 0.7725 | 0.7675 |
1.3506 | 43.98 | 733 | 1.3716 | 0.7675 | 0.08 | 0.765 | 0.7675 | 0.77 | 0.7675 |
1.3506 | 45.0 | 750 | 1.3829 | 0.77 | 0.08 | 0.755 | 0.765 | 0.7675 | 0.7675 |
1.3506 | 45.96 | 766 | 1.3846 | 0.7675 | 0.08 | 0.7625 | 0.765 | 0.765 | 0.765 |
1.3506 | 46.98 | 783 | 1.3797 | 0.7725 | 0.08 | 0.765 | 0.7675 | 0.77 | 0.765 |
1.3506 | 48.0 | 800 | 1.3792 | 0.77 | 0.08 | 0.7675 | 0.7675 | 0.7675 | 0.7675 |
1.3506 | 48.96 | 816 | 1.3771 | 0.775 | 0.08 | 0.765 | 0.77 | 0.77 | 0.77 |
1.3506 | 49.98 | 833 | 1.3806 | 0.7725 | 0.08 | 0.765 | 0.77 | 0.7725 | 0.775 |
1.3506 | 51.0 | 850 | 1.3842 | 0.77 | 0.08 | 0.765 | 0.7675 | 0.77 | 0.77 |
1.3506 | 51.96 | 866 | 1.3854 | 0.7675 | 0.08 | 0.765 | 0.77 | 0.7675 | 0.765 |
1.3506 | 52.98 | 883 | 1.3869 | 0.77 | 0.08 | 0.765 | 0.77 | 0.77 | 0.77 |
1.3506 | 54.0 | 900 | 1.3899 | 0.7675 | 0.08 | 0.765 | 0.7725 | 0.7675 | 0.765 |
1.3506 | 54.96 | 916 | 1.3916 | 0.7675 | 0.08 | 0.765 | 0.7725 | 0.7675 | 0.765 |
1.3506 | 55.98 | 933 | 1.3915 | 0.7675 | 0.08 | 0.765 | 0.7675 | 0.7675 | 0.765 |
1.3506 | 57.0 | 950 | 1.3922 | 0.7675 | 0.08 | 0.7675 | 0.7675 | 0.7675 | 0.765 |
1.3506 | 57.6 | 960 | 1.3922 | 0.7675 | 0.08 | 0.7675 | 0.7675 | 0.7675 | 0.765 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3