Hugging Face's logo

Lilya
/
distilbert-base-uncased-ner-invoiceSenderRecipient-all-inv-26-12

Token Classification
Transformers
PyTorch
distilbert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
Lilya commited on
Commit
f24343b
·
1 Parent(s): a133264

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -25
README.md CHANGED
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model was trained from scratch on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.0253
22
- - Precision: 0.8377
23
- - Recall: 0.8893
24
- - F1: 0.8627
25
- - Accuracy: 0.9911
26
 
27
  ## Model description
28
 
@@ -47,29 +47,47 @@ The following hyperparameters were used during training:
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
- - num_epochs: 1
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
55
- |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
56
- | 0.0301 | 0.06 | 500 | 0.0312 | 0.8199 | 0.8282 | 0.8240 | 0.9892 |
57
- | 0.0279 | 0.11 | 1000 | 0.0307 | 0.8106 | 0.8602 | 0.8347 | 0.9894 |
58
- | 0.0253 | 0.17 | 1500 | 0.0317 | 0.8272 | 0.8282 | 0.8277 | 0.9890 |
59
- | 0.025 | 0.23 | 2000 | 0.0311 | 0.8056 | 0.8699 | 0.8365 | 0.9894 |
60
- | 0.0249 | 0.28 | 2500 | 0.0327 | 0.7927 | 0.8798 | 0.8340 | 0.9888 |
61
- | 0.0234 | 0.34 | 3000 | 0.0306 | 0.7948 | 0.8862 | 0.8380 | 0.9894 |
62
- | 0.0232 | 0.4 | 3500 | 0.0305 | 0.8172 | 0.8803 | 0.8476 | 0.9900 |
63
- | 0.0235 | 0.46 | 4000 | 0.0295 | 0.8289 | 0.8666 | 0.8473 | 0.9902 |
64
- | 0.0232 | 0.51 | 4500 | 0.0291 | 0.8048 | 0.8866 | 0.8437 | 0.9899 |
65
- | 0.0228 | 0.57 | 5000 | 0.0289 | 0.8234 | 0.8839 | 0.8525 | 0.9904 |
66
- | 0.0246 | 0.63 | 5500 | 0.0292 | 0.8129 | 0.8921 | 0.8506 | 0.9901 |
67
- | 0.024 | 0.68 | 6000 | 0.0271 | 0.8199 | 0.8908 | 0.8538 | 0.9905 |
68
- | 0.0285 | 0.74 | 6500 | 0.0267 | 0.8262 | 0.8913 | 0.8575 | 0.9906 |
69
- | 0.0275 | 0.8 | 7000 | 0.0260 | 0.8325 | 0.8887 | 0.8597 | 0.9909 |
70
- | 0.0274 | 0.85 | 7500 | 0.0261 | 0.8311 | 0.8924 | 0.8607 | 0.9909 |
71
- | 0.0262 | 0.91 | 8000 | 0.0254 | 0.8359 | 0.8906 | 0.8624 | 0.9910 |
72
- | 0.027 | 0.97 | 8500 | 0.0253 | 0.8377 | 0.8893 | 0.8627 | 0.9911 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
73
 
74
 
75
  ### Framework versions
 
18
 
19
  This model was trained from scratch on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.0240
22
+ - Precision: 0.8517
23
+ - Recall: 0.9006
24
+ - F1: 0.8755
25
+ - Accuracy: 0.9918
26
 
27
  ## Model description
28
 
 
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
+ - num_epochs: 2
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
55
+ |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
56
+ | 0.0302 | 0.06 | 500 | 0.0312 | 0.8199 | 0.8274 | 0.8236 | 0.9892 |
57
+ | 0.028 | 0.11 | 1000 | 0.0308 | 0.8108 | 0.8591 | 0.8343 | 0.9894 |
58
+ | 0.0255 | 0.17 | 1500 | 0.0319 | 0.8278 | 0.8226 | 0.8252 | 0.9890 |
59
+ | 0.0253 | 0.23 | 2000 | 0.0314 | 0.8046 | 0.8674 | 0.8348 | 0.9893 |
60
+ | 0.0253 | 0.28 | 2500 | 0.0329 | 0.7914 | 0.8783 | 0.8326 | 0.9887 |
61
+ | 0.0239 | 0.34 | 3000 | 0.0309 | 0.7907 | 0.8871 | 0.8361 | 0.9893 |
62
+ | 0.0238 | 0.4 | 3500 | 0.0313 | 0.8109 | 0.8822 | 0.8450 | 0.9898 |
63
+ | 0.0242 | 0.46 | 4000 | 0.0292 | 0.8290 | 0.8646 | 0.8464 | 0.9902 |
64
+ | 0.0239 | 0.51 | 4500 | 0.0302 | 0.7938 | 0.8859 | 0.8373 | 0.9895 |
65
+ | 0.0237 | 0.57 | 5000 | 0.0291 | 0.8246 | 0.8795 | 0.8512 | 0.9903 |
66
+ | 0.0254 | 0.63 | 5500 | 0.0296 | 0.8160 | 0.8884 | 0.8507 | 0.9901 |
67
+ | 0.0248 | 0.68 | 6000 | 0.0270 | 0.8269 | 0.8845 | 0.8547 | 0.9906 |
68
+ | 0.029 | 0.74 | 6500 | 0.0271 | 0.8283 | 0.8874 | 0.8568 | 0.9906 |
69
+ | 0.0277 | 0.8 | 7000 | 0.0259 | 0.8374 | 0.8823 | 0.8593 | 0.9909 |
70
+ | 0.0276 | 0.85 | 7500 | 0.0264 | 0.8317 | 0.8930 | 0.8612 | 0.9909 |
71
+ | 0.0263 | 0.91 | 8000 | 0.0252 | 0.8412 | 0.8901 | 0.8650 | 0.9911 |
72
+ | 0.0271 | 0.97 | 8500 | 0.0247 | 0.8531 | 0.8756 | 0.8642 | 0.9913 |
73
+ | 0.0242 | 1.02 | 9000 | 0.0256 | 0.8459 | 0.8909 | 0.8678 | 0.9913 |
74
+ | 0.022 | 1.08 | 9500 | 0.0262 | 0.8310 | 0.9000 | 0.8641 | 0.9910 |
75
+ | 0.0212 | 1.14 | 10000 | 0.0251 | 0.8581 | 0.8780 | 0.8679 | 0.9914 |
76
+ | 0.0215 | 1.19 | 10500 | 0.0255 | 0.8441 | 0.8952 | 0.8689 | 0.9914 |
77
+ | 0.0209 | 1.25 | 11000 | 0.0253 | 0.8378 | 0.8982 | 0.8669 | 0.9913 |
78
+ | 0.021 | 1.31 | 11500 | 0.0253 | 0.8358 | 0.9049 | 0.8690 | 0.9913 |
79
+ | 0.0211 | 1.37 | 12000 | 0.0252 | 0.8437 | 0.8989 | 0.8704 | 0.9915 |
80
+ | 0.0205 | 1.42 | 12500 | 0.0249 | 0.8464 | 0.8980 | 0.8714 | 0.9916 |
81
+ | 0.0206 | 1.48 | 13000 | 0.0247 | 0.8440 | 0.8973 | 0.8698 | 0.9916 |
82
+ | 0.02 | 1.54 | 13500 | 0.0246 | 0.8528 | 0.8939 | 0.8729 | 0.9916 |
83
+ | 0.0208 | 1.59 | 14000 | 0.0249 | 0.8397 | 0.9063 | 0.8718 | 0.9915 |
84
+ | 0.0205 | 1.65 | 14500 | 0.0241 | 0.8549 | 0.8932 | 0.8736 | 0.9917 |
85
+ | 0.0204 | 1.71 | 15000 | 0.0241 | 0.8534 | 0.8976 | 0.8749 | 0.9918 |
86
+ | 0.0196 | 1.76 | 15500 | 0.0246 | 0.8464 | 0.9038 | 0.8741 | 0.9917 |
87
+ | 0.0202 | 1.82 | 16000 | 0.0239 | 0.8514 | 0.8990 | 0.8746 | 0.9918 |
88
+ | 0.0197 | 1.88 | 16500 | 0.0242 | 0.8494 | 0.9008 | 0.8744 | 0.9917 |
89
+ | 0.0198 | 1.93 | 17000 | 0.0240 | 0.8514 | 0.9005 | 0.8752 | 0.9918 |
90
+ | 0.0202 | 1.99 | 17500 | 0.0240 | 0.8517 | 0.9007 | 0.8755 | 0.9918 |
91
 
92
 
93
  ### Framework versions