Hugging Face's logo

HueyNemud
/
icdar23-entrydetector_plaintext

Token Classification
Transformers
PyTorch
camembert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
HueyNemud commited on
Commit
bb054c3
•
1 Parent(s): 6c37c8d

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -19
README.md CHANGED
@@ -13,12 +13,12 @@ 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: 0.0215
17
- - Ebegin: {'precision': 0.9807757166947724, 'recall': 0.9686875416389074, 'f1': 0.9746941511647393, 'number': 3002}
18
- - Eend: {'precision': 0.9847457627118644, 'recall': 0.9683333333333334, 'f1': 0.9764705882352941, 'number': 3000}
19
- - Overall Precision: 0.9828
20
- - Overall Recall: 0.9685
21
- - Overall F1: 0.9756
22
  - Overall Accuracy: 0.9954
23
 
24
  ## Model description
@@ -50,19 +50,13 @@ 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 | 0.0344 | 0.9832 | 0.9771 | 0.9801 | 0.9960 |
54
- | 0.1472 | 0.14 | 600 | 0.0254 | 0.9909 | 0.9672 | 0.9789 | 0.9958 |
55
- | 0.1472 | 0.21 | 900 | 0.0200 | 0.9892 | 0.9734 | 0.9813 | 0.9963 |
56
- | 0.026 | 0.29 | 1200 | 0.0176 | 0.9840 | 0.9782 | 0.9811 | 0.9961 |
57
- | 0.0179 | 0.36 | 1500 | 0.0163 | 0.9952 | 0.9662 | 0.9805 | 0.9961 |
58
- | 0.0179 | 0.43 | 1800 | 0.0154 | 0.9942 | 0.9706 | 0.9823 | 0.9964 |
59
- | 0.0141 | 0.5 | 2100 | 0.0143 | 0.9910 | 0.9733 | 0.9821 | 0.9964 |
60
- | 0.0141 | 0.57 | 2400 | 0.0249 | 0.9917 | 0.9488 | 0.9697 | 0.9939 |
61
- | 0.0158 | 0.64 | 2700 | 0.0130 | 0.9980 | 0.9713 | 0.9844 | 0.9968 |
62
- | 0.0143 | 0.72 | 3000 | 0.0128 | 0.9888 | 0.9759 | 0.9823 | 0.9964 |
63
- | 0.0143 | 0.79 | 3300 | 0.0145 | 0.9888 | 0.9760 | 0.9823 | 0.9964 |
64
- | 0.0124 | 0.86 | 3600 | 0.0144 | 0.9857 | 0.9802 | 0.9829 | 0.9965 |
65
- | 0.0124 | 0.93 | 3900 | 0.0141 | 0.9871 | 0.9804 | 0.9837 | 0.9967 |
66
 
67
 
68
  ### 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: 0.0191
17
+ - Ebegin: {'precision': 0.993127147766323, 'recall': 0.9626915389740173, 'f1': 0.9776725304465493, 'number': 3002}
18
+ - Eend: {'precision': 0.9910313901345291, 'recall': 0.9576666666666667, 'f1': 0.9740634005763689, 'number': 3000}
19
+ - Overall Precision: 0.9921
20
+ - Overall Recall: 0.9602
21
+ - Overall F1: 0.9759
22
  - Overall Accuracy: 0.9954
23
 
24
  ## Model description
 
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
53
+ | No log | 0.07 | 300 | 0.0312 | 0.9804 | 0.9791 | 0.9797 | 0.9960 |
54
+ | 0.1423 | 0.14 | 600 | 0.0232 | 0.9920 | 0.9670 | 0.9793 | 0.9958 |
55
+ | 0.1423 | 0.21 | 900 | 0.0164 | 0.9959 | 0.9696 | 0.9825 | 0.9965 |
56
+ | 0.0242 | 0.29 | 1200 | 0.0174 | 0.9850 | 0.9744 | 0.9797 | 0.9959 |
57
+ | 0.0169 | 0.36 | 1500 | 0.0165 | 0.9913 | 0.9696 | 0.9803 | 0.9960 |
58
+ | 0.0169 | 0.43 | 1800 | 0.0168 | 0.9908 | 0.9715 | 0.9810 | 0.9962 |
59
+ | 0.0153 | 0.5 | 2100 | 0.0164 | 0.9884 | 0.9702 | 0.9793 | 0.9960 |
 
 
 
 
 
 
60
 
61
 
62
  ### Framework versions