update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
model-index:
|
| 5 |
+
- name: icdar23-entrydetector_labelledtext_breaks_indents_left_diff_right_ref
|
| 6 |
+
results: []
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 10 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
+
|
| 12 |
+
# icdar23-entrydetector_labelledtext_breaks_indents_left_diff_right_ref
|
| 13 |
+
|
| 14 |
+
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.
|
| 15 |
+
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.2515
|
| 17 |
+
- Act: {'precision': 0.8046783625730994, 'recall': 0.9017038007863696, 'f1': 0.8504326328800988, 'number': 1526}
|
| 18 |
+
- Cardinal: {'precision': 0.9451428571428572, 'recall': 0.9538638985005767, 'f1': 0.9494833524684271, 'number': 2601}
|
| 19 |
+
- Ebegin: {'precision': 0.9940431868950111, 'recall': 0.9910913140311804, 'f1': 0.9925650557620818, 'number': 2694}
|
| 20 |
+
- Eend: {'precision': 0.9988751406074241, 'recall': 0.9859363434492968, 'f1': 0.9923635686347551, 'number': 2702}
|
| 21 |
+
- Ft: {'precision': 0.2, 'recall': 0.2857142857142857, 'f1': 0.23529411764705882, 'number': 21}
|
| 22 |
+
- Loc: {'precision': 0.9071332436069987, 'recall': 0.935072142064373, 'f1': 0.9208908320808854, 'number': 3604}
|
| 23 |
+
- Per: {'precision': 0.9300651354130957, 'recall': 0.9345504650361695, 'f1': 0.9323024054982818, 'number': 2903}
|
| 24 |
+
- Titre: {'precision': 0.5234042553191489, 'recall': 0.82, 'f1': 0.6389610389610388, 'number': 150}
|
| 25 |
+
- Overall Precision: 0.9287
|
| 26 |
+
- Overall Recall: 0.9507
|
| 27 |
+
- Overall F1: 0.9396
|
| 28 |
+
- Overall Accuracy: 0.9459
|
| 29 |
+
|
| 30 |
+
## Model description
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Intended uses & limitations
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training and evaluation data
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Training procedure
|
| 43 |
+
|
| 44 |
+
### Training hyperparameters
|
| 45 |
+
|
| 46 |
+
The following hyperparameters were used during training:
|
| 47 |
+
- learning_rate: 0.0001
|
| 48 |
+
- train_batch_size: 2
|
| 49 |
+
- eval_batch_size: 2
|
| 50 |
+
- seed: 42
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: linear
|
| 53 |
+
- training_steps: 7500
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 58 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 59 |
+
| No log | 0.07 | 300 | 0.2219 | 0.8682 | 0.9280 | 0.8971 | 0.9525 |
|
| 60 |
+
| 0.4302 | 0.14 | 600 | 0.1580 | 0.9429 | 0.9488 | 0.9458 | 0.9658 |
|
| 61 |
+
| 0.4302 | 0.21 | 900 | 0.1575 | 0.9398 | 0.9429 | 0.9413 | 0.9597 |
|
| 62 |
+
| 0.1819 | 0.29 | 1200 | 0.1236 | 0.9447 | 0.9542 | 0.9495 | 0.9686 |
|
| 63 |
+
| 0.1537 | 0.36 | 1500 | 0.1312 | 0.9543 | 0.9486 | 0.9514 | 0.9670 |
|
| 64 |
+
| 0.1537 | 0.43 | 1800 | 0.1337 | 0.9487 | 0.9559 | 0.9523 | 0.9679 |
|
| 65 |
+
| 0.119 | 0.5 | 2100 | 0.1198 | 0.9554 | 0.9556 | 0.9555 | 0.9702 |
|
| 66 |
+
| 0.119 | 0.57 | 2400 | 0.1128 | 0.9467 | 0.9641 | 0.9553 | 0.9707 |
|
| 67 |
+
| 0.1098 | 0.64 | 2700 | 0.1215 | 0.9528 | 0.9607 | 0.9567 | 0.9713 |
|
| 68 |
+
| 0.1118 | 0.72 | 3000 | 0.1099 | 0.9482 | 0.9635 | 0.9558 | 0.9711 |
|
| 69 |
+
| 0.1118 | 0.79 | 3300 | 0.1140 | 0.9541 | 0.9684 | 0.9612 | 0.9727 |
|
| 70 |
+
| 0.094 | 0.86 | 3600 | 0.0969 | 0.9581 | 0.9654 | 0.9617 | 0.9748 |
|
| 71 |
+
| 0.094 | 0.93 | 3900 | 0.1089 | 0.9564 | 0.9664 | 0.9614 | 0.9755 |
|
| 72 |
+
| 0.0895 | 1.0 | 4200 | 0.1158 | 0.9574 | 0.9662 | 0.9618 | 0.9746 |
|
| 73 |
+
| 0.0626 | 1.07 | 4500 | 0.1072 | 0.9479 | 0.9709 | 0.9593 | 0.9747 |
|
| 74 |
+
| 0.0626 | 1.14 | 4800 | 0.1060 | 0.9549 | 0.9682 | 0.9615 | 0.9735 |
|
| 75 |
+
| 0.0474 | 1.22 | 5100 | 0.1172 | 0.9462 | 0.9718 | 0.9588 | 0.9723 |
|
| 76 |
+
| 0.0474 | 1.29 | 5400 | 0.1019 | 0.9550 | 0.9698 | 0.9624 | 0.9764 |
|
| 77 |
+
| 0.0554 | 1.36 | 5700 | 0.1086 | 0.9473 | 0.9700 | 0.9585 | 0.9737 |
|
| 78 |
+
| 0.0416 | 1.43 | 6000 | 0.1175 | 0.9514 | 0.9714 | 0.9613 | 0.9737 |
|
| 79 |
+
| 0.0416 | 1.5 | 6300 | 0.1143 | 0.9536 | 0.9718 | 0.9626 | 0.9742 |
|
| 80 |
+
| 0.0514 | 1.57 | 6600 | 0.1113 | 0.9618 | 0.9679 | 0.9648 | 0.9749 |
|
| 81 |
+
| 0.0514 | 1.65 | 6900 | 0.1084 | 0.9595 | 0.9709 | 0.9652 | 0.9762 |
|
| 82 |
+
| 0.0377 | 1.72 | 7200 | 0.1102 | 0.9601 | 0.9706 | 0.9653 | 0.9759 |
|
| 83 |
+
| 0.0437 | 1.79 | 7500 | 0.1123 | 0.9585 | 0.9710 | 0.9647 | 0.9757 |
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
### Framework versions
|
| 87 |
+
|
| 88 |
+
- Transformers 4.26.1
|
| 89 |
+
- Pytorch 1.13.1+cu116
|
| 90 |
+
- Datasets 2.9.0
|
| 91 |
+
- Tokenizers 0.13.2
|