metadata
tags:
- generated_from_trainer
model-index:
- name: icdar23-entrydetector_plaintext_breaks
results: []
icdar23-entrydetector_plaintext_breaks
This model is a fine-tuned version of HueyNemud/das22-10-camembert_pretrained on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0171
- Ebegin: {'precision': 0.9908381404818459, 'recall': 0.9726848767488341, 'f1': 0.9816775928727518, 'number': 3002}
- Eend: {'precision': 0.9856379425517702, 'recall': 0.9836666666666667, 'f1': 0.9846513179846514, 'number': 3000}
- Overall Precision: 0.9882
- Overall Recall: 0.9782
- Overall F1: 0.9832
- Overall Accuracy: 0.9971
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: 6000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.07 | 300 | 0.0317 | 0.9815 | 0.9778 | 0.9796 | 0.9965 |
0.1664 | 0.14 | 600 | 0.0166 | 0.9892 | 0.9815 | 0.9853 | 0.9972 |
0.1664 | 0.21 | 900 | 0.0158 | 0.9773 | 0.9861 | 0.9817 | 0.9965 |
0.0198 | 0.29 | 1200 | 0.0127 | 0.9834 | 0.9864 | 0.9849 | 0.9971 |
0.0152 | 0.36 | 1500 | 0.0126 | 0.9860 | 0.9842 | 0.9851 | 0.9971 |
0.0152 | 0.43 | 1800 | 0.0118 | 0.9961 | 0.9733 | 0.9846 | 0.9971 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2