Hugging Face's logo

mp-02
/
layoutlmv3-base-cord-sroie

Token Classification
Transformers
TensorBoard
Safetensors
layoutlmv3
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
mp-02 commited on
Commit
3233879
·
verified ·
1 Parent(s): af8c981

End of training

Browse files
Files changed (6) hide show
  1. README.md +30 -25
  2. all_results.json +8 -8
  3. predict_results.json +8 -8
  4. predictions.txt +0 -0
  5. runs/Nov17_22-14-34_347f825b15ea/events.out.tfevents.1731881675.347f825b15ea.3217.1 +2 -2
  6. runs/Nov17_22-14-34_347f825b15ea/events.out.tfevents.1731884167.347f825b15ea.3217.2 +3 -0
README.md CHANGED
@@ -22,16 +22,16 @@ model-index:
22
  metrics:
23
  - name: Precision
24
  type: precision
25
- value: 0.9399720800372267
26
  - name: Recall
27
  type: recall
28
- value: 0.9465791940018744
29
  - name: F1
30
  type: f1
31
- value: 0.9432640672425869
32
  - name: Accuracy
33
  type: accuracy
34
- value: 0.9813340410474168
35
  ---
36
 
37
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -41,11 +41,11 @@ should probably proofread and complete it, then remove this comment. -->
41
 
42
  This model is a fine-tuned version of [layoutlmv3](https://huggingface.co/layoutlmv3) on the mp-02/cord-sroie dataset.
43
  It achieves the following results on the evaluation set:
44
- - Loss: 0.0970
45
- - Precision: 0.9400
46
- - Recall: 0.9466
47
- - F1: 0.9433
48
- - Accuracy: 0.9813
49
 
50
  ## Model description
51
 
@@ -65,29 +65,34 @@ More information needed
65
 
66
  The following hyperparameters were used during training:
67
  - learning_rate: 2e-05
68
- - train_batch_size: 16
69
- - eval_batch_size: 16
70
  - seed: 42
71
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
72
  - lr_scheduler_type: linear
73
  - training_steps: 4000
74
 
75
  ### Training results
76
 
77
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
78
- |:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
79
- | No log | 2.2222 | 100 | 0.3258 | 0.8171 | 0.7685 | 0.7921 | 0.9363 |
80
- | No log | 4.4444 | 200 | 0.1516 | 0.9078 | 0.8946 | 0.9011 | 0.9694 |
81
- | No log | 6.6667 | 300 | 0.1085 | 0.9315 | 0.9175 | 0.9245 | 0.9761 |
82
- | No log | 8.8889 | 400 | 0.1000 | 0.9382 | 0.9456 | 0.9419 | 0.9817 |
83
- | 0.4015 | 11.1111 | 500 | 0.0970 | 0.9400 | 0.9466 | 0.9433 | 0.9813 |
84
- | 0.4015 | 13.3333 | 600 | 0.1064 | 0.9505 | 0.9358 | 0.9431 | 0.9814 |
85
- | 0.4015 | 15.5556 | 700 | 0.1095 | 0.9465 | 0.9372 | 0.9418 | 0.9812 |
 
 
 
 
 
86
 
87
 
88
  ### Framework versions
89
 
90
- - Transformers 4.44.2
91
- - Pytorch 2.4.0+cu118
92
- - Datasets 2.21.0
93
- - Tokenizers 0.19.1
 
22
  metrics:
23
  - name: Precision
24
  type: precision
25
+ value: 0.9036656236030398
26
  - name: Recall
27
  type: recall
28
+ value: 0.9578298981284056
29
  - name: F1
30
  type: f1
31
+ value: 0.9299597469810236
32
  - name: Accuracy
33
  type: accuracy
34
+ value: 0.9736783204261605
35
  ---
36
 
37
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
41
 
42
  This model is a fine-tuned version of [layoutlmv3](https://huggingface.co/layoutlmv3) on the mp-02/cord-sroie dataset.
43
  It achieves the following results on the evaluation set:
44
+ - Loss: 0.0967
45
+ - Precision: 0.9037
46
+ - Recall: 0.9578
47
+ - F1: 0.9300
48
+ - Accuracy: 0.9737
49
 
50
  ## Model description
51
 
 
65
 
66
  The following hyperparameters were used during training:
67
  - learning_rate: 2e-05
68
+ - train_batch_size: 8
69
+ - eval_batch_size: 8
70
  - seed: 42
71
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
72
  - lr_scheduler_type: linear
73
  - training_steps: 4000
74
 
75
  ### Training results
76
 
77
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
78
+ |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
79
+ | No log | 0.7937 | 100 | 0.4387 | 0.6226 | 0.5785 | 0.5998 | 0.9037 |
80
+ | No log | 1.5873 | 200 | 0.2236 | 0.8925 | 0.8439 | 0.8675 | 0.9562 |
81
+ | No log | 2.3810 | 300 | 0.1342 | 0.9127 | 0.8965 | 0.9045 | 0.9692 |
82
+ | No log | 3.1746 | 400 | 0.1054 | 0.9119 | 0.9273 | 0.9195 | 0.9735 |
83
+ | 0.6635 | 3.9683 | 500 | 0.1341 | 0.8555 | 0.9495 | 0.9001 | 0.9630 |
84
+ | 0.6635 | 4.7619 | 600 | 0.1060 | 0.9059 | 0.9493 | 0.9271 | 0.9739 |
85
+ | 0.6635 | 5.5556 | 700 | 0.1066 | 0.9080 | 0.9420 | 0.9247 | 0.9738 |
86
+ | 0.6635 | 6.3492 | 800 | 0.1008 | 0.9078 | 0.9564 | 0.9315 | 0.9746 |
87
+ | 0.6635 | 7.1429 | 900 | 0.0988 | 0.9086 | 0.9517 | 0.9296 | 0.9738 |
88
+ | 0.0995 | 7.9365 | 1000 | 0.0967 | 0.9037 | 0.9578 | 0.9300 | 0.9737 |
89
+ | 0.0995 | 8.7302 | 1100 | 0.1224 | 0.8777 | 0.9642 | 0.9189 | 0.9690 |
90
+ | 0.0995 | 9.5238 | 1200 | 0.1263 | 0.8879 | 0.9536 | 0.9196 | 0.9694 |
91
 
92
 
93
  ### Framework versions
94
 
95
+ - Transformers 4.46.2
96
+ - Pytorch 2.5.1+cu121
97
+ - Datasets 3.1.0
98
+ - Tokenizers 0.20.3
all_results.json CHANGED
@@ -1,10 +1,10 @@
1
  {
2
- "predict_accuracy": 0.9703861414884352,
3
- "predict_f1": 0.923658709524935,
4
- "predict_loss": 0.14045372605323792,
5
- "predict_precision": 0.9067285382830627,
6
- "predict_recall": 0.941233140655106,
7
- "predict_runtime": 12.1468,
8
- "predict_samples_per_second": 11.279,
9
- "predict_steps_per_second": 0.741
10
  }
 
1
  {
2
+ "predict_accuracy": 0.9776242335499222,
3
+ "predict_f1": 0.9420884632922936,
4
+ "predict_loss": 0.09829691052436829,
5
+ "predict_precision": 0.9247985675917636,
6
+ "predict_recall": 0.9600371747211895,
7
+ "predict_runtime": 37.955,
8
+ "predict_samples_per_second": 7.193,
9
+ "predict_steps_per_second": 0.922
10
  }
predict_results.json CHANGED
@@ -1,10 +1,10 @@
1
  {
2
- "predict_accuracy": 0.9703861414884352,
3
- "predict_f1": 0.923658709524935,
4
- "predict_loss": 0.14045372605323792,
5
- "predict_precision": 0.9067285382830627,
6
- "predict_recall": 0.941233140655106,
7
- "predict_runtime": 12.1468,
8
- "predict_samples_per_second": 11.279,
9
- "predict_steps_per_second": 0.741
10
  }
 
1
  {
2
+ "predict_accuracy": 0.9776242335499222,
3
+ "predict_f1": 0.9420884632922936,
4
+ "predict_loss": 0.09829691052436829,
5
+ "predict_precision": 0.9247985675917636,
6
+ "predict_recall": 0.9600371747211895,
7
+ "predict_runtime": 37.955,
8
+ "predict_samples_per_second": 7.193,
9
+ "predict_steps_per_second": 0.922
10
  }
predictions.txt CHANGED
The diff for this file is too large to render. See raw diff
 
runs/Nov17_22-14-34_347f825b15ea/events.out.tfevents.1731881675.347f825b15ea.3217.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6b6073a4262ac8fcb04c9cae175e9e94d58c26a59da99198456bb7ffe28c345
3
- size 13069
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c2d30b02608afd522ef2602b1c2e4e8db78f49e0af43655dbdf776f6c1d88f7
3
+ size 14367
runs/Nov17_22-14-34_347f825b15ea/events.out.tfevents.1731884167.347f825b15ea.3217.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ca9fb8e32b34c0662eb44a0bc52d1012613c242673bd8602a8b993cffa9b0b7
3
+ size 560