update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
model-index:
|
| 5 |
+
- name: icdar23-entrydetector_jointlabelledtext_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_jointlabelledtext_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: 0.2882
|
| 17 |
+
- Act: {'precision': 0.829136690647482, 'recall': 0.9062909567496723, 'f1': 0.8659987476518473, 'number': 1526}
|
| 18 |
+
- Cardinal: {'precision': 0.969980506822612, 'recall': 0.97339593114241, 'f1': 0.9716852177309119, 'number': 2556}
|
| 19 |
+
- Cardinal+i-eend: {'precision': 1.0, 'recall': 0.32456140350877194, 'f1': 0.490066225165563, 'number': 114}
|
| 20 |
+
- Ft: {'precision': 0.1935483870967742, 'recall': 0.2857142857142857, 'f1': 0.23076923076923075, 'number': 21}
|
| 21 |
+
- Loc: {'precision': 0.9216652971788551, 'recall': 0.9349819394276188, 'f1': 0.9282758620689655, 'number': 3599}
|
| 22 |
+
- Loc+i-eend: {'precision': 0.75, 'recall': 0.44680851063829785, 'f1': 0.56, 'number': 47}
|
| 23 |
+
- Per: {'precision': 0.9322283609576427, 'recall': 0.9264275256222547, 'f1': 0.9293188911327336, 'number': 2732}
|
| 24 |
+
- Per+i-ebegin: {'precision': 0.9908045977011494, 'recall': 0.9923254029163469, 'f1': 0.9915644171779141, 'number': 2606}
|
| 25 |
+
- Titre: {'precision': 0.6735751295336787, 'recall': 0.8666666666666667, 'f1': 0.7580174927113703, 'number': 150}
|
| 26 |
+
- Overall Precision: 0.9295
|
| 27 |
+
- Overall Recall: 0.9398
|
| 28 |
+
- Overall F1: 0.9346
|
| 29 |
+
- Overall Accuracy: 0.9445
|
| 30 |
+
|
| 31 |
+
## Model description
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Intended uses & limitations
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training and evaluation data
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Training procedure
|
| 44 |
+
|
| 45 |
+
### Training hyperparameters
|
| 46 |
+
|
| 47 |
+
The following hyperparameters were used during training:
|
| 48 |
+
- learning_rate: 0.0001
|
| 49 |
+
- train_batch_size: 2
|
| 50 |
+
- eval_batch_size: 2
|
| 51 |
+
- seed: 42
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: linear
|
| 54 |
+
- training_steps: 7500
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 60 |
+
| No log | 0.07 | 300 | 0.2325 | 0.8550 | 0.9146 | 0.8838 | 0.9556 |
|
| 61 |
+
| 0.5551 | 0.14 | 600 | 0.1765 | 0.9370 | 0.9372 | 0.9371 | 0.9647 |
|
| 62 |
+
| 0.5551 | 0.21 | 900 | 0.1533 | 0.9306 | 0.9369 | 0.9337 | 0.9628 |
|
| 63 |
+
| 0.2064 | 0.29 | 1200 | 0.1283 | 0.9446 | 0.9487 | 0.9467 | 0.9712 |
|
| 64 |
+
| 0.1584 | 0.36 | 1500 | 0.1497 | 0.9456 | 0.9447 | 0.9452 | 0.9664 |
|
| 65 |
+
| 0.1584 | 0.43 | 1800 | 0.1406 | 0.9357 | 0.9544 | 0.9450 | 0.9679 |
|
| 66 |
+
| 0.1313 | 0.5 | 2100 | 0.1303 | 0.9339 | 0.9530 | 0.9433 | 0.9686 |
|
| 67 |
+
| 0.1313 | 0.57 | 2400 | 0.1208 | 0.9518 | 0.9571 | 0.9545 | 0.9742 |
|
| 68 |
+
| 0.1186 | 0.64 | 2700 | 0.1229 | 0.9459 | 0.9563 | 0.9511 | 0.9728 |
|
| 69 |
+
| 0.1157 | 0.72 | 3000 | 0.1053 | 0.9522 | 0.9573 | 0.9547 | 0.9739 |
|
| 70 |
+
| 0.1157 | 0.79 | 3300 | 0.1051 | 0.9456 | 0.9566 | 0.9511 | 0.9740 |
|
| 71 |
+
| 0.0899 | 0.86 | 3600 | 0.1083 | 0.9504 | 0.9571 | 0.9537 | 0.9740 |
|
| 72 |
+
| 0.0899 | 0.93 | 3900 | 0.1032 | 0.9487 | 0.9589 | 0.9538 | 0.9741 |
|
| 73 |
+
| 0.0946 | 1.0 | 4200 | 0.1106 | 0.9519 | 0.9571 | 0.9545 | 0.9745 |
|
| 74 |
+
| 0.0621 | 1.07 | 4500 | 0.1051 | 0.9431 | 0.9720 | 0.9573 | 0.9756 |
|
| 75 |
+
| 0.0621 | 1.14 | 4800 | 0.1019 | 0.9489 | 0.9655 | 0.9571 | 0.9747 |
|
| 76 |
+
| 0.0504 | 1.22 | 5100 | 0.1334 | 0.9452 | 0.9685 | 0.9567 | 0.9722 |
|
| 77 |
+
| 0.0504 | 1.29 | 5400 | 0.1175 | 0.9526 | 0.9625 | 0.9575 | 0.9745 |
|
| 78 |
+
| 0.0478 | 1.36 | 5700 | 0.1166 | 0.9480 | 0.9680 | 0.9579 | 0.9748 |
|
| 79 |
+
| 0.042 | 1.43 | 6000 | 0.1126 | 0.9463 | 0.9659 | 0.9560 | 0.9744 |
|
| 80 |
+
| 0.042 | 1.5 | 6300 | 0.1143 | 0.9427 | 0.9712 | 0.9567 | 0.9738 |
|
| 81 |
+
| 0.0512 | 1.57 | 6600 | 0.1119 | 0.9558 | 0.9615 | 0.9586 | 0.9750 |
|
| 82 |
+
| 0.0512 | 1.65 | 6900 | 0.1159 | 0.9548 | 0.9663 | 0.9605 | 0.9758 |
|
| 83 |
+
| 0.0381 | 1.72 | 7200 | 0.1159 | 0.9595 | 0.9650 | 0.9623 | 0.9768 |
|
| 84 |
+
| 0.0455 | 1.79 | 7500 | 0.1161 | 0.9570 | 0.9661 | 0.9615 | 0.9763 |
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
### Framework versions
|
| 88 |
+
|
| 89 |
+
- Transformers 4.26.1
|
| 90 |
+
- Pytorch 1.13.1+cu116
|
| 91 |
+
- Datasets 2.9.0
|
| 92 |
+
- Tokenizers 0.13.2
|