|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: icdar23-entrydetector_plaintext_breaks_indents_left_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_plaintext_breaks_indents_left_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: 0.0079 |
|
- Ebegin: {'precision': 0.9973414356247626, 'recall': 0.9875893192929672, 'f1': 0.9924414210128495, 'number': 2659} |
|
- Eend: {'precision': 0.9980966882375333, 'recall': 0.9798206278026906, 'f1': 0.9888742221384123, 'number': 2676} |
|
- Overall Precision: 0.9977 |
|
- Overall Recall: 0.9837 |
|
- Overall F1: 0.9907 |
|
- Overall Accuracy: 0.9984 |
|
|
|
## 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 | 0.0341 | 0.9873 | 0.9698 | 0.9785 | 0.9966 | |
|
| 0.173 | 0.14 | 600 | 0.0140 | 0.9895 | 0.9899 | 0.9897 | 0.9982 | |
|
| 0.173 | 0.21 | 900 | 0.0135 | 0.9796 | 0.9884 | 0.9840 | 0.9973 | |
|
| 0.0216 | 0.29 | 1200 | 0.0087 | 0.9938 | 0.9901 | 0.9920 | 0.9986 | |
|
| 0.0138 | 0.36 | 1500 | 0.0061 | 0.9884 | 0.9938 | 0.9911 | 0.9984 | |
|
| 0.0138 | 0.43 | 1800 | 0.0060 | 0.9938 | 0.9919 | 0.9929 | 0.9987 | |
|
| 0.0083 | 0.5 | 2100 | 0.0058 | 0.9963 | 0.9909 | 0.9935 | 0.9989 | |
|
| 0.0083 | 0.57 | 2400 | 0.0064 | 0.9972 | 0.9913 | 0.9942 | 0.9990 | |
|
| 0.0092 | 0.64 | 2700 | 0.0083 | 0.9881 | 0.9947 | 0.9914 | 0.9985 | |
|
| 0.0087 | 0.72 | 3000 | 0.0057 | 0.9924 | 0.9934 | 0.9929 | 0.9987 | |
|
| 0.0087 | 0.79 | 3300 | 0.0044 | 0.9925 | 0.9927 | 0.9926 | 0.9987 | |
|
| 0.0066 | 0.86 | 3600 | 0.0049 | 0.9948 | 0.9917 | 0.9932 | 0.9988 | |
|
| 0.0066 | 0.93 | 3900 | 0.0082 | 0.9886 | 0.9916 | 0.9901 | 0.9982 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|
