Hugging Face's logo

TomasFAV
/
Layoutlmv3InvoiceCzechV0

Token Classification
Transformers
TensorBoard
Safetensors
layoutlmv3
Generated from Trainer
invoice-processing
information-extraction
czech-language
document-ai
layout-aware-model
multimodal-model
synthetic-data
Model card Files
xet
 Metrics Community
TomasFAV commited on
Commit
e52cfdc
·
verified ·
1 Parent(s): 9ad9c66

Model save

Browse files
Files changed (3) hide show
  1. README.md +20 -30
  2. model.safetensors +1 -1
  3. runs/Mar07_20-17-45_6bfa1b237cbf/events.out.tfevents.1772917028.6bfa1b237cbf.4077.1 +3 -0
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 [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.0649
25
- - Precision: 0.8992
26
- - Recall: 0.9155
27
- - F1: 0.9072
28
- - Accuracy: 0.9862
29
 
30
  ## Model description
31
 
@@ -45,44 +45,34 @@ More information needed
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 1e-05
48
- - train_batch_size: 16
49
  - eval_batch_size: 2
50
  - seed: 42
51
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
  - lr_scheduler_type: linear
53
- - lr_scheduler_warmup_ratio: 0.1
54
- - num_epochs: 20
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
60
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
61
- | No log | 1.0 | 43 | 1.5391 | 0.0 | 0.0 | 0.0 | 0.8323 |
62
- | No log | 2.0 | 86 | 1.0087 | 0.0 | 0.0 | 0.0 | 0.8323 |
63
- | No log | 3.0 | 129 | 0.6795 | 0.0739 | 0.0338 | 0.0464 | 0.8375 |
64
- | No log | 4.0 | 172 | 0.5549 | 0.2447 | 0.2802 | 0.2613 | 0.8690 |
65
- | No log | 5.0 | 215 | 0.4249 | 0.4758 | 0.5821 | 0.5236 | 0.9183 |
66
- | No log | 6.0 | 258 | 0.3194 | 0.5484 | 0.6836 | 0.6086 | 0.9374 |
67
- | No log | 7.0 | 301 | 0.2431 | 0.6235 | 0.7379 | 0.6759 | 0.9502 |
68
- | No log | 8.0 | 344 | 0.1812 | 0.7525 | 0.8116 | 0.7809 | 0.9673 |
69
- | No log | 9.0 | 387 | 0.1548 | 0.7991 | 0.8357 | 0.8170 | 0.9705 |
70
- | No log | 10.0 | 430 | 0.1248 | 0.8182 | 0.8478 | 0.8327 | 0.9758 |
71
- | No log | 11.0 | 473 | 0.1104 | 0.8451 | 0.8696 | 0.8571 | 0.9786 |
72
- | 0.64 | 12.0 | 516 | 0.0965 | 0.8506 | 0.8732 | 0.8617 | 0.9804 |
73
- | 0.64 | 13.0 | 559 | 0.0909 | 0.8688 | 0.8877 | 0.8781 | 0.9820 |
74
- | 0.64 | 14.0 | 602 | 0.0819 | 0.8826 | 0.8986 | 0.8905 | 0.9837 |
75
- | 0.64 | 15.0 | 645 | 0.0779 | 0.8897 | 0.9058 | 0.8977 | 0.9850 |
76
- | 0.64 | 16.0 | 688 | 0.0734 | 0.8942 | 0.9082 | 0.9011 | 0.9852 |
77
- | 0.64 | 17.0 | 731 | 0.0700 | 0.8955 | 0.9106 | 0.9030 | 0.9856 |
78
- | 0.64 | 18.0 | 774 | 0.0666 | 0.8988 | 0.9118 | 0.9053 | 0.9861 |
79
- | 0.64 | 19.0 | 817 | 0.0655 | 0.8992 | 0.9155 | 0.9072 | 0.9862 |
80
- | 0.64 | 20.0 | 860 | 0.0649 | 0.8992 | 0.9155 | 0.9072 | 0.9862 |
81
 
82
 
83
  ### Framework versions
84
 
85
- - Transformers 4.57.6
86
- - Pytorch 2.9.0+cu126
87
  - Datasets 4.0.0
88
  - Tokenizers 0.22.2
 
21
 
22
  This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.2200
25
+ - Precision: 0.4808
26
+ - Recall: 0.7469
27
+ - F1: 0.5850
28
+ - Accuracy: 0.9620
29
 
30
  ## Model description
31
 
 
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 1e-05
48
+ - train_batch_size: 8
49
  - eval_batch_size: 2
50
  - seed: 42
51
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
  - lr_scheduler_type: linear
53
+ - lr_scheduler_warmup_steps: 0.1
54
+ - num_epochs: 10
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
60
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
61
+ | No log | 1.0 | 150 | 0.2635 | 0.0 | 0.0 | 0.0 | 0.9556 |
62
+ | No log | 2.0 | 300 | 0.1821 | 0.3558 | 0.5329 | 0.4267 | 0.9571 |
63
+ | No log | 3.0 | 450 | 0.1810 | 0.4170 | 0.6461 | 0.5069 | 0.9575 |
64
+ | 0.5190 | 4.0 | 600 | 0.2093 | 0.3975 | 0.6461 | 0.4922 | 0.9549 |
65
+ | 0.5190 | 5.0 | 750 | 0.1810 | 0.4432 | 0.6584 | 0.5298 | 0.9617 |
66
+ | 0.5190 | 6.0 | 900 | 0.2108 | 0.4622 | 0.7160 | 0.5617 | 0.9614 |
67
+ | 0.0314 | 7.0 | 1050 | 0.2065 | 0.4612 | 0.7222 | 0.5630 | 0.9602 |
68
+ | 0.0314 | 8.0 | 1200 | 0.2199 | 0.4808 | 0.7469 | 0.5850 | 0.9620 |
69
+ | 0.0314 | 9.0 | 1350 | 0.2267 | 0.4681 | 0.7551 | 0.5780 | 0.9602 |
70
+ | 0.0134 | 10.0 | 1500 | 0.2220 | 0.4735 | 0.7551 | 0.5821 | 0.9608 |
 
 
 
 
 
 
 
 
 
 
71
 
72
 
73
  ### Framework versions
74
 
75
+ - Transformers 5.0.0
76
+ - Pytorch 2.10.0+cu128
77
  - Datasets 4.0.0
78
  - Tokenizers 0.22.2
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4182b6939f3d02fe8ca5f498001809305af31fa6d68596f72fac8da3938eb63a
3
  size 503791932
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:375cc120b4c0280bc0b7234917d6f01d775baf79bce644958f855d33590a443c
3
  size 503791932
runs/Mar07_20-17-45_6bfa1b237cbf/events.out.tfevents.1772917028.6bfa1b237cbf.4077.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:430d566716b90753366fca1a657ac163f6db278b56fa898e8ab438e48ee8184e
3
+ size 560