|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: icdar23-entrydetector_texttokens_breaks_indents_left_diff_right_ref |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# icdar23-entrydetector_texttokens_breaks_indents_left_diff_right_ref |
|
|
|
This model is a fine-tuned version of [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/HueyNemud/das22-10-camembert_pretrained) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0444 |
|
- Ebegin: {'precision': 0.9864814119414195, 'recall': 0.9751299183370453, 'f1': 0.9807728206085495, 'number': 2694} |
|
- Eend: {'precision': 0.9851024208566108, 'recall': 0.9789045151739453, 'f1': 0.9819936885093744, 'number': 2702} |
|
- Overall Precision: 0.9858 |
|
- Overall Recall: 0.9770 |
|
- Overall F1: 0.9814 |
|
- Overall Accuracy: 0.9868 |
|
|
|
## 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: 0.0001 |
|
- 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: 7500 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 0.07 | 300 | 0.0977 | 0.9752 | 0.9604 | 0.9677 | 0.9788 | |
|
| 0.31 | 0.14 | 600 | 0.0922 | 0.9910 | 0.9415 | 0.9656 | 0.9779 | |
|
| 0.31 | 0.21 | 900 | 0.0628 | 0.9926 | 0.9534 | 0.9726 | 0.9823 | |
|
| 0.1143 | 0.29 | 1200 | 0.0570 | 0.9715 | 0.9802 | 0.9759 | 0.9838 | |
|
| 0.0878 | 0.36 | 1500 | 0.0393 | 0.9914 | 0.9731 | 0.9822 | 0.9885 | |
|
| 0.0878 | 0.43 | 1800 | 0.0437 | 0.9825 | 0.9819 | 0.9822 | 0.9883 | |
|
| 0.0884 | 0.5 | 2100 | 0.0296 | 0.9908 | 0.9861 | 0.9885 | 0.9924 | |
|
| 0.0884 | 0.57 | 2400 | 0.0340 | 0.9913 | 0.9837 | 0.9875 | 0.9918 | |
|
| 0.0898 | 0.64 | 2700 | 0.0294 | 0.9833 | 0.9932 | 0.9882 | 0.9923 | |
|
| 0.066 | 0.72 | 3000 | 0.0369 | 0.9853 | 0.9849 | 0.9851 | 0.9904 | |
|
| 0.066 | 0.79 | 3300 | 0.0245 | 0.9892 | 0.9889 | 0.9890 | 0.9928 | |
|
| 0.0575 | 0.86 | 3600 | 0.0230 | 0.9879 | 0.9924 | 0.9901 | 0.9936 | |
|
| 0.0575 | 0.93 | 3900 | 0.0282 | 0.9865 | 0.9831 | 0.9848 | 0.9908 | |
|
| 0.064 | 1.0 | 4200 | 0.0244 | 0.9945 | 0.9822 | 0.9883 | 0.9923 | |
|
| 0.0626 | 1.07 | 4500 | 0.0203 | 0.9929 | 0.9880 | 0.9905 | 0.9937 | |
|
| 0.0626 | 1.14 | 4800 | 0.0198 | 0.9920 | 0.9891 | 0.9905 | 0.9937 | |
|
| 0.0419 | 1.22 | 5100 | 0.0219 | 0.9895 | 0.9878 | 0.9886 | 0.9925 | |
|
| 0.0419 | 1.29 | 5400 | 0.0235 | 0.9890 | 0.9876 | 0.9883 | 0.9923 | |
|
| 0.0564 | 1.36 | 5700 | 0.0212 | 0.9935 | 0.9880 | 0.9908 | 0.9939 | |
|
| 0.0427 | 1.43 | 6000 | 0.0238 | 0.9934 | 0.9839 | 0.9886 | 0.9925 | |
|
| 0.0427 | 1.5 | 6300 | 0.0193 | 0.9862 | 0.9920 | 0.9891 | 0.9928 | |
|
| 0.0501 | 1.57 | 6600 | 0.0212 | 0.9919 | 0.9885 | 0.9902 | 0.9935 | |
|
| 0.0501 | 1.65 | 6900 | 0.0225 | 0.9911 | 0.9880 | 0.9896 | 0.9931 | |
|
| 0.0488 | 1.72 | 7200 | 0.0212 | 0.9904 | 0.9892 | 0.9898 | 0.9933 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|