|
--- |
|
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: 0.0474 |
|
- Ebegin: {'precision': 0.9822681935722202, 'recall': 0.9870081662954714, 'f1': 0.9846324754675061, 'number': 2694} |
|
- Eend: {'precision': 0.9790132547864506, 'recall': 0.9840858623242043, 'f1': 0.9815430047988187, 'number': 2702} |
|
- Overall Precision: 0.9806 |
|
- Overall Recall: 0.9855 |
|
- Overall F1: 0.9831 |
|
- Overall Accuracy: 0.9880 |
|
|
|
## 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 | 0.0916 | 0.9560 | 0.9910 | 0.9731 | 0.9821 | |
|
| 0.32 | 0.14 | 600 | 0.0893 | 0.9919 | 0.9482 | 0.9695 | 0.9803 | |
|
| 0.32 | 0.21 | 900 | 0.0772 | 0.9903 | 0.9419 | 0.9655 | 0.9778 | |
|
| 0.114 | 0.29 | 1200 | 0.0442 | 0.9811 | 0.9798 | 0.9805 | 0.9871 | |
|
| 0.0983 | 0.36 | 1500 | 0.0351 | 0.9907 | 0.9826 | 0.9866 | 0.9912 | |
|
| 0.0983 | 0.43 | 1800 | 0.0325 | 0.9917 | 0.9856 | 0.9887 | 0.9926 | |
|
| 0.0869 | 0.5 | 2100 | 0.0237 | 0.9905 | 0.9938 | 0.9921 | 0.9948 | |
|
| 0.0869 | 0.57 | 2400 | 0.0316 | 0.9890 | 0.9870 | 0.9880 | 0.9921 | |
|
| 0.0844 | 0.64 | 2700 | 0.0271 | 0.9863 | 0.9892 | 0.9877 | 0.9919 | |
|
| 0.0794 | 0.72 | 3000 | 0.0287 | 0.9872 | 0.9901 | 0.9887 | 0.9926 | |
|
| 0.0794 | 0.79 | 3300 | 0.0281 | 0.9862 | 0.9891 | 0.9876 | 0.9918 | |
|
| 0.0561 | 0.86 | 3600 | 0.0259 | 0.9883 | 0.9934 | 0.9908 | 0.9939 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|
