Hugging Face's logo

mp-02
/
layoutlmv3-finetuned-funsd

PyTorch
TensorBoard
Safetensors
layoutlmv3
Generated from Trainer
Eval Results
Model card Files Metrics Community
mp-02 commited on
Commit
8cb6f0d
•
1 Parent(s): fd7a3f0

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -18
README.md CHANGED
@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.8330
23
- - Precision: 0.9046
24
- - Recall: 0.9105
25
- - F1: 0.9076
26
- - Accuracy: 0.8536
27
 
28
  ## Model description
29
 
@@ -42,28 +42,30 @@ More information needed
42
  ### Training hyperparameters
43
 
44
  The following hyperparameters were used during training:
45
- - learning_rate: 5e-05
46
  - train_batch_size: 4
47
- - eval_batch_size: 16
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
- - training_steps: 500
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
- | No log | 1.32 | 50 | 0.6163 | 0.8088 | 0.8965 | 0.8504 | 0.8088 |
58
- | No log | 2.63 | 100 | 0.5416 | 0.8037 | 0.868 | 0.8346 | 0.8134 |
59
- | No log | 3.95 | 150 | 0.5572 | 0.8446 | 0.8885 | 0.8660 | 0.8385 |
60
- | No log | 5.26 | 200 | 0.7317 | 0.8458 | 0.8555 | 0.8506 | 0.8124 |
61
- | No log | 6.58 | 250 | 0.7220 | 0.8877 | 0.8935 | 0.8906 | 0.8385 |
62
- | No log | 7.89 | 300 | 0.8070 | 0.8778 | 0.9055 | 0.8915 | 0.8436 |
63
- | No log | 9.21 | 350 | 0.7895 | 0.8969 | 0.913 | 0.9049 | 0.8477 |
64
- | No log | 10.53 | 400 | 0.8168 | 0.8935 | 0.889 | 0.8912 | 0.8412 |
65
- | No log | 11.84 | 450 | 0.8233 | 0.8955 | 0.917 | 0.9061 | 0.8521 |
66
- | 0.2564 | 13.16 | 500 | 0.8330 | 0.9046 | 0.9105 | 0.9076 | 0.8536 |
 
 
67
 
68
 
69
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.7162
23
+ - Precision: 0.7822
24
+ - Recall: 0.835
25
+ - F1: 0.8077
26
+ - Accuracy: 0.7967
27
 
28
  ## Model description
29
 
 
42
  ### Training hyperparameters
43
 
44
  The following hyperparameters were used during training:
45
+ - learning_rate: 1e-05
46
  - train_batch_size: 4
47
+ - eval_batch_size: 14
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
+ - training_steps: 120
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | No log | 0.26 | 10 | 1.7343 | 0.0730 | 0.0865 | 0.0792 | 0.3227 |
58
+ | No log | 0.53 | 20 | 1.4897 | 0.1517 | 0.1465 | 0.1490 | 0.4537 |
59
+ | No log | 0.79 | 30 | 1.2939 | 0.3717 | 0.4785 | 0.4184 | 0.5962 |
60
+ | No log | 1.05 | 40 | 1.1471 | 0.5035 | 0.5725 | 0.5358 | 0.6368 |
61
+ | No log | 1.32 | 50 | 1.0022 | 0.6310 | 0.6525 | 0.6416 | 0.7383 |
62
+ | No log | 1.58 | 60 | 0.9246 | 0.6739 | 0.713 | 0.6929 | 0.7382 |
63
+ | No log | 1.84 | 70 | 0.8455 | 0.7159 | 0.7825 | 0.7477 | 0.7544 |
64
+ | No log | 2.11 | 80 | 0.7844 | 0.7477 | 0.8105 | 0.7778 | 0.7817 |
65
+ | No log | 2.37 | 90 | 0.7947 | 0.7476 | 0.819 | 0.7817 | 0.7456 |
66
+ | No log | 2.63 | 100 | 0.7346 | 0.7824 | 0.834 | 0.8074 | 0.7959 |
67
+ | No log | 2.89 | 110 | 0.7200 | 0.7870 | 0.837 | 0.8112 | 0.7954 |
68
+ | No log | 3.16 | 120 | 0.7162 | 0.7822 | 0.835 | 0.8077 | 0.7967 |
69
 
70
 
71
  ### Framework versions