File size: 3,581 Bytes
962fc35 e78d099 962fc35 c0a11e1 962fc35 e78d099 962fc35 c0a11e1 962fc35 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
---
tags:
- generated_from_trainer
model-index:
- name: icdar23-entrydetector_plaintext_breaks_indents_left_ref_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_plaintext_breaks_indents_left_ref_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: 0.0033
- Ebegin: {'precision': 0.9935849056603774, 'recall': 0.9902218879277924, 'f1': 0.9919005462422301, 'number': 2659}
- Eend: {'precision': 0.996606334841629, 'recall': 0.9876681614349776, 'f1': 0.9921171171171171, 'number': 2676}
- Overall Precision: 0.9951
- Overall Recall: 0.9889
- Overall F1: 0.9920
- Overall Accuracy: 0.9989
## 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 | 0.0262 | 0.9758 | 0.9906 | 0.9832 | 0.9973 |
| 0.1599 | 0.14 | 600 | 0.0116 | 0.9866 | 0.9919 | 0.9892 | 0.9982 |
| 0.1599 | 0.21 | 900 | 0.0111 | 0.9907 | 0.9856 | 0.9882 | 0.9980 |
| 0.0162 | 0.29 | 1200 | 0.0096 | 0.9813 | 0.9966 | 0.9889 | 0.9981 |
| 0.0099 | 0.36 | 1500 | 0.0060 | 0.9820 | 0.9955 | 0.9887 | 0.9982 |
| 0.0099 | 0.43 | 1800 | 0.0046 | 0.9925 | 0.9934 | 0.9929 | 0.9988 |
| 0.0074 | 0.5 | 2100 | 0.0057 | 0.9961 | 0.9880 | 0.9920 | 0.9987 |
| 0.0074 | 0.57 | 2400 | 0.0039 | 0.9911 | 0.9953 | 0.9932 | 0.9988 |
| 0.0072 | 0.64 | 2700 | 0.0075 | 0.9842 | 0.9949 | 0.9895 | 0.9982 |
| 0.0061 | 0.72 | 3000 | 0.0040 | 0.9906 | 0.9963 | 0.9934 | 0.9989 |
| 0.0061 | 0.79 | 3300 | 0.0034 | 0.9955 | 0.9936 | 0.9946 | 0.9991 |
| 0.005 | 0.86 | 3600 | 0.0034 | 0.9933 | 0.9946 | 0.9939 | 0.9990 |
| 0.005 | 0.93 | 3900 | 0.0047 | 0.9847 | 0.9976 | 0.9911 | 0.9985 |
| 0.0041 | 1.0 | 4200 | 0.0031 | 0.9972 | 0.9936 | 0.9954 | 0.9992 |
| 0.0031 | 1.07 | 4500 | 0.0030 | 0.9967 | 0.9945 | 0.9956 | 0.9992 |
| 0.0031 | 1.14 | 4800 | 0.0032 | 0.9966 | 0.9938 | 0.9952 | 0.9992 |
| 0.003 | 1.22 | 5100 | 0.0029 | 0.9960 | 0.9939 | 0.9949 | 0.9991 |
| 0.003 | 1.29 | 5400 | 0.0030 | 0.9935 | 0.9947 | 0.9941 | 0.9990 |
| 0.0023 | 1.36 | 5700 | 0.0028 | 0.9973 | 0.9933 | 0.9953 | 0.9992 |
| 0.0027 | 1.43 | 6000 | 0.0029 | 0.9968 | 0.9936 | 0.9952 | 0.9992 |
### Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2
|