smitbutle/document-data-extraction-layoutlmv3

1---2license: cc-by-nc-sa-4.03base_model: microsoft/layoutlmv3-base4tags:5- generated_from_trainer6datasets:7- generated8metrics:9- precision10- recall11- f112- accuracy13model-index:14- name: document-data-extraction-layoutlmv315  results:16  - task:17      name: Token Classification18      type: token-classification19    dataset:20      name: generated21      type: generated22      config: sroie23      split: test24      args: sroie25    metrics:26    - name: Precision27      type: precision28      value: 1.029    - name: Recall30      type: recall31      value: 1.032    - name: F133      type: f134      value: 1.035    - name: Accuracy36      type: accuracy37      value: 1.038---39 40<!-- This model card has been generated automatically according to the information the Trainer had access to. You41should probably proofread and complete it, then remove this comment. -->42 43# document-data-extraction-layoutlmv344 45This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the generated dataset.46It achieves the following results on the evaluation set:47- Loss: 0.001548- Precision: 1.049- Recall: 1.050- F1: 1.051- Accuracy: 1.052 53## Model description54 55More information needed56 57## Intended uses & limitations58 59More information needed60 61## Training and evaluation data62 63More information needed64 65## Training procedure66 67### Training hyperparameters68 69The following hyperparameters were used during training:70- learning_rate: 1e-0571- train_batch_size: 172- eval_batch_size: 173- seed: 4274- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0875- lr_scheduler_type: linear76- training_steps: 200077 78### Training results79 80| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |81|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|82| No log        | 1.0   | 100  | 0.1114          | 0.95      | 0.9635 | 0.9567 | 0.9947   |83| No log        | 2.0   | 200  | 0.0286          | 0.972     | 0.9858 | 0.9789 | 0.9971   |84| No log        | 3.0   | 300  | 0.0184          | 0.972     | 0.9858 | 0.9789 | 0.9971   |85| No log        | 4.0   | 400  | 0.0163          | 0.972     | 0.9858 | 0.9789 | 0.9971   |86| 0.1385        | 5.0   | 500  | 0.0141          | 0.972     | 0.9858 | 0.9789 | 0.9971   |87| 0.1385        | 6.0   | 600  | 0.0123          | 0.972     | 0.9858 | 0.9789 | 0.9971   |88| 0.1385        | 7.0   | 700  | 0.0122          | 0.972     | 0.9858 | 0.9789 | 0.9971   |89| 0.1385        | 8.0   | 800  | 0.0108          | 0.972     | 0.9858 | 0.9789 | 0.9971   |90| 0.1385        | 9.0   | 900  | 0.0104          | 0.972     | 0.9858 | 0.9789 | 0.9971   |91| 0.0119        | 10.0  | 1000 | 0.0113          | 0.972     | 0.9858 | 0.9789 | 0.9971   |92| 0.0119        | 11.0  | 1100 | 0.0080          | 0.974     | 0.9878 | 0.9809 | 0.9973   |93| 0.0119        | 12.0  | 1200 | 0.0089          | 0.9856    | 0.9736 | 0.9796 | 0.9973   |94| 0.0119        | 13.0  | 1300 | 0.0034          | 0.9959    | 0.9959 | 0.9959 | 0.9994   |95| 0.0119        | 14.0  | 1400 | 0.0037          | 0.9980    | 0.9939 | 0.9959 | 0.9994   |96| 0.006         | 15.0  | 1500 | 0.0024          | 0.9960    | 0.9980 | 0.9970 | 0.9996   |97| 0.006         | 16.0  | 1600 | 0.0019          | 0.9980    | 1.0    | 0.9990 | 0.9998   |98| 0.006         | 17.0  | 1700 | 0.0022          | 0.9960    | 0.9980 | 0.9970 | 0.9996   |99| 0.006         | 18.0  | 1800 | 0.0017          | 1.0       | 1.0    | 1.0    | 1.0      |100| 0.006         | 19.0  | 1900 | 0.0015          | 1.0       | 1.0    | 1.0    | 1.0      |101| 0.0027        | 20.0  | 2000 | 0.0015          | 1.0       | 1.0    | 1.0    | 1.0      |102 103 104### Framework versions105 106- Transformers 4.36.0.dev0107- Pytorch 2.1.1+cu121108- Datasets 2.15.0109- Tokenizers 0.15.0110