Hugging Face's logo

NLPmonster
/
layoutlmv3-for-complete-receipt-understanding

Token Classification
Transformers
TensorBoard
Safetensors
layoutlmv3
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
NLPmonster commited on
Commit
224c1e3
·
verified ·
1 Parent(s): 61dc2fb

layoutlmv3-for-complete-receipt-understanding

Browse files
Files changed (4) hide show
  1. README.md +26 -25
  2. model.safetensors +1 -1
  3. runs/Oct10_17-53-11_58282926b10a/events.out.tfevents.1728582796.58282926b10a.36703.0 +3 -0
  4. training_args.bin +1 -1
README.md CHANGED
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [NLPmonster/layoutlmv3-for-receipt-understanding](https://huggingface.co/NLPmonster/layoutlmv3-for-receipt-understanding) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.2598
25
- - Precision: 0.6952
26
- - Recall: 0.7574
27
- - F1: 0.7249
28
- - Accuracy: 0.9105
29
 
30
  ## Model description
31
 
@@ -56,30 +56,31 @@ The following hyperparameters were used during training:
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
58
  |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
59
- | 0.8058 | 0.4425 | 50 | 0.5154 | 0.5535 | 0.6755 | 0.6084 | 0.8422 |
60
- | 0.4654 | 0.8850 | 100 | 0.3750 | 0.6294 | 0.7001 | 0.6629 | 0.8711 |
61
- | 0.3668 | 1.3274 | 150 | 0.3623 | 0.6992 | 0.6536 | 0.6756 | 0.8917 |
62
- | 0.3432 | 1.7699 | 200 | 0.3644 | 0.6114 | 0.7046 | 0.6547 | 0.8755 |
63
- | 0.4204 | 2.2124 | 250 | 0.4515 | 0.6089 | 0.6978 | 0.6503 | 0.8735 |
64
- | 0.2933 | 2.6549 | 300 | 0.3410 | 0.6308 | 0.7198 | 0.6724 | 0.8842 |
65
- | 0.2681 | 3.0973 | 350 | 0.3544 | 0.6205 | 0.7284 | 0.6701 | 0.8783 |
66
- | 0.23 | 3.5398 | 400 | 0.3059 | 0.6475 | 0.7331 | 0.6876 | 0.8905 |
67
- | 0.2286 | 3.9823 | 450 | 0.2810 | 0.6715 | 0.7243 | 0.6969 | 0.8999 |
68
- | 0.1924 | 4.4248 | 500 | 0.2556 | 0.7504 | 0.7336 | 0.7419 | 0.9182 |
69
- | 0.1826 | 4.8673 | 550 | 0.2894 | 0.7217 | 0.6779 | 0.6991 | 0.9067 |
70
- | 0.167 | 5.3097 | 600 | 0.2721 | 0.7117 | 0.7524 | 0.7314 | 0.9129 |
71
- | 0.1338 | 5.7522 | 650 | 0.2596 | 0.7226 | 0.7497 | 0.7359 | 0.9159 |
72
- | 0.142 | 6.1947 | 700 | 0.2501 | 0.7492 | 0.7422 | 0.7457 | 0.9214 |
73
- | 0.1169 | 6.6372 | 750 | 0.2661 | 0.6870 | 0.7512 | 0.7177 | 0.9071 |
74
- | 0.1145 | 7.0796 | 800 | 0.2707 | 0.6913 | 0.7539 | 0.7212 | 0.9082 |
75
- | 0.1094 | 7.5221 | 850 | 0.2717 | 0.6889 | 0.7530 | 0.7195 | 0.9067 |
76
- | 0.0993 | 7.9646 | 900 | 0.2574 | 0.6918 | 0.7545 | 0.7218 | 0.9103 |
77
- | 0.0928 | 8.4071 | 950 | 0.2605 | 0.6947 | 0.7594 | 0.7256 | 0.9104 |
78
- | 0.0985 | 8.8496 | 1000 | 0.2598 | 0.6952 | 0.7574 | 0.7249 | 0.9105 |
79
 
80
 
81
  ### Framework versions
82
 
83
  - Transformers 4.44.2
84
  - Pytorch 2.4.1+cu121
 
85
  - Tokenizers 0.19.1
 
21
 
22
  This model is a fine-tuned version of [NLPmonster/layoutlmv3-for-receipt-understanding](https://huggingface.co/NLPmonster/layoutlmv3-for-receipt-understanding) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.4459
25
+ - Precision: 0.8739
26
+ - Recall: 0.8808
27
+ - F1: 0.8773
28
+ - Accuracy: 0.8751
29
 
30
  ## Model description
31
 
 
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
58
  |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
59
+ | 0.923 | 0.4425 | 50 | 0.6403 | 0.8015 | 0.8231 | 0.8121 | 0.8043 |
60
+ | 0.6865 | 0.8850 | 100 | 0.5970 | 0.8097 | 0.8356 | 0.8224 | 0.8132 |
61
+ | 0.5204 | 1.3274 | 150 | 0.5144 | 0.8409 | 0.8406 | 0.8407 | 0.8375 |
62
+ | 0.4726 | 1.7699 | 200 | 0.5400 | 0.8273 | 0.8420 | 0.8346 | 0.8292 |
63
+ | 0.417 | 2.2124 | 250 | 0.5776 | 0.8107 | 0.8413 | 0.8257 | 0.8185 |
64
+ | 0.3753 | 2.6549 | 300 | 0.4914 | 0.8358 | 0.8388 | 0.8373 | 0.8369 |
65
+ | 0.3314 | 3.0973 | 350 | 0.5010 | 0.8265 | 0.8477 | 0.8370 | 0.8326 |
66
+ | 0.2861 | 3.5398 | 400 | 0.4890 | 0.8486 | 0.8537 | 0.8512 | 0.8492 |
67
+ | 0.2834 | 3.9823 | 450 | 0.5026 | 0.8396 | 0.8680 | 0.8535 | 0.8462 |
68
+ | 0.2226 | 4.4248 | 500 | 0.4843 | 0.8476 | 0.8569 | 0.8522 | 0.8508 |
69
+ | 0.2192 | 4.8673 | 550 | 0.4660 | 0.8527 | 0.8676 | 0.8601 | 0.856 |
70
+ | 0.183 | 5.3097 | 600 | 0.4524 | 0.8592 | 0.8730 | 0.8660 | 0.8634 |
71
+ | 0.1718 | 5.7522 | 650 | 0.4594 | 0.8579 | 0.8747 | 0.8663 | 0.8606 |
72
+ | 0.1612 | 6.1947 | 700 | 0.4486 | 0.8663 | 0.8737 | 0.8700 | 0.8668 |
73
+ | 0.136 | 6.6372 | 750 | 0.4589 | 0.8613 | 0.8662 | 0.8637 | 0.8594 |
74
+ | 0.1261 | 7.0796 | 800 | 0.4282 | 0.8734 | 0.8740 | 0.8737 | 0.8708 |
75
+ | 0.111 | 7.5221 | 850 | 0.4591 | 0.8692 | 0.8730 | 0.8711 | 0.8689 |
76
+ | 0.1024 | 7.9646 | 900 | 0.4467 | 0.8720 | 0.8822 | 0.8771 | 0.8729 |
77
+ | 0.0822 | 8.4071 | 950 | 0.4414 | 0.8711 | 0.8801 | 0.8756 | 0.8729 |
78
+ | 0.0816 | 8.8496 | 1000 | 0.4459 | 0.8739 | 0.8808 | 0.8773 | 0.8751 |
79
 
80
 
81
  ### Framework versions
82
 
83
  - Transformers 4.44.2
84
  - Pytorch 2.4.1+cu121
85
+ - Datasets 3.0.1
86
  - Tokenizers 0.19.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6f555ff99ba60784247476b443a8e2a26df8b3c14240522b177bfe8d5056d035
3
  size 503825792
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72de31fc8df1625eabf8baec3891baa4cd8b96cdbb7f00118b2e272576bbc1dc
3
  size 503825792
runs/Oct10_17-53-11_58282926b10a/events.out.tfevents.1728582796.58282926b10a.36703.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d952688011f8dbe3f823ee2332ce5a89976095044a1eb3f50f19d0060a742931
3
+ size 21864
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7a2bfbae64bc6a2fd3d6bd0257e12c0b57d30f6729c1be6586dea001125e777
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4152bf1122bd8372170593b69d18ba26bce0a663a6ee1b317f2588d64e3ad8d5
3
  size 5240