Hugging Face's logo

HueyNemud
/
icdar23-entrydetector_plaintext_breaks

Token Classification
Transformers
PyTorch
camembert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
HueyNemud commited on
Commit
9953ad8
•
1 Parent(s): ad479f4

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -24
README.md CHANGED
@@ -13,13 +13,13 @@ should probably proofread and complete it, then remove this comment. -->
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: 0.0084
17
- - Ebegin: {'precision': 0.9966113181972213, 'recall': 0.9796802131912059, 'f1': 0.9880732403830002, 'number': 3002}
18
- - Eend: {'precision': 0.9870603848706039, 'recall': 0.9916666666666667, 'f1': 0.9893581642833389, 'number': 3000}
19
- - Overall Precision: 0.9918
20
- - Overall Recall: 0.9857
21
- - Overall F1: 0.9887
22
- - Overall Accuracy: 0.9979
23
 
24
  ## Model description
25
 
@@ -50,23 +50,12 @@ The following hyperparameters were used during training:
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
53
- | No log | 0.07 | 300 | 0.0332 | 0.9838 | 0.9824 | 0.9831 | 0.9968 |
54
- | 0.1693 | 0.14 | 600 | 0.0242 | 0.9760 | 0.9806 | 0.9783 | 0.9958 |
55
- | 0.1693 | 0.21 | 900 | 0.0162 | 0.9863 | 0.9836 | 0.9849 | 0.9971 |
56
- | 0.024 | 0.29 | 1200 | 0.0172 | 0.9895 | 0.9775 | 0.9835 | 0.9969 |
57
- | 0.0229 | 0.36 | 1500 | 0.0143 | 0.9895 | 0.9822 | 0.9858 | 0.9973 |
58
- | 0.0229 | 0.43 | 1800 | 0.0123 | 0.9926 | 0.9809 | 0.9867 | 0.9975 |
59
- | 0.0143 | 0.5 | 2100 | 0.0120 | 0.9938 | 0.9759 | 0.9848 | 0.9971 |
60
- | 0.0143 | 0.57 | 2400 | 0.0136 | 0.9917 | 0.9809 | 0.9862 | 0.9974 |
61
- | 0.016 | 0.64 | 2700 | 0.0112 | 0.9945 | 0.9801 | 0.9872 | 0.9976 |
62
- | 0.0112 | 0.72 | 3000 | 0.0106 | 0.9835 | 0.9869 | 0.9852 | 0.9972 |
63
- | 0.0112 | 0.79 | 3300 | 0.0098 | 0.9899 | 0.9864 | 0.9882 | 0.9977 |
64
- | 0.0115 | 0.86 | 3600 | 0.0092 | 0.9897 | 0.9867 | 0.9882 | 0.9978 |
65
- | 0.0115 | 0.93 | 3900 | 0.0093 | 0.9911 | 0.9856 | 0.9884 | 0.9978 |
66
- | 0.012 | 1.0 | 4200 | 0.0115 | 0.9890 | 0.9855 | 0.9872 | 0.9976 |
67
- | 0.0087 | 1.07 | 4500 | 0.0095 | 0.9870 | 0.9869 | 0.9870 | 0.9975 |
68
- | 0.0087 | 1.14 | 4800 | 0.0092 | 0.9863 | 0.9862 | 0.9862 | 0.9974 |
69
- | 0.0089 | 1.22 | 5100 | 0.0090 | 0.9907 | 0.9855 | 0.9880 | 0.9977 |
70
 
71
 
72
  ### Framework versions
 
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: 0.0171
17
+ - Ebegin: {'precision': 0.9908381404818459, 'recall': 0.9726848767488341, 'f1': 0.9816775928727518, 'number': 3002}
18
+ - Eend: {'precision': 0.9856379425517702, 'recall': 0.9836666666666667, 'f1': 0.9846513179846514, 'number': 3000}
19
+ - Overall Precision: 0.9882
20
+ - Overall Recall: 0.9782
21
+ - Overall F1: 0.9832
22
+ - Overall Accuracy: 0.9971
23
 
24
  ## Model description
25
 
 
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
53
+ | No log | 0.07 | 300 | 0.0317 | 0.9815 | 0.9778 | 0.9796 | 0.9965 |
54
+ | 0.1664 | 0.14 | 600 | 0.0166 | 0.9892 | 0.9815 | 0.9853 | 0.9972 |
55
+ | 0.1664 | 0.21 | 900 | 0.0158 | 0.9773 | 0.9861 | 0.9817 | 0.9965 |
56
+ | 0.0198 | 0.29 | 1200 | 0.0127 | 0.9834 | 0.9864 | 0.9849 | 0.9971 |
57
+ | 0.0152 | 0.36 | 1500 | 0.0126 | 0.9860 | 0.9842 | 0.9851 | 0.9971 |
58
+ | 0.0152 | 0.43 | 1800 | 0.0118 | 0.9961 | 0.9733 | 0.9846 | 0.9971 |
 
 
 
 
 
 
 
 
 
 
 
59
 
60
 
61
  ### Framework versions