Hugging Face's logo

tiennvcs
/
layoutlmv2-large-uncased-finetuned-vi-infovqa

Document Question Answering Transformers PyTorch TensorBoard layoutlmv2 generated_from_trainer Inference Endpoints
Model card Files Metrics Community
2
tiennvcs commited on
Commit
6db2513
•
1 Parent(s): 07dca22

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +41 -17
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [microsoft/layoutlmv2-large-uncased](https://huggingface.co/microsoft/layoutlmv2-large-uncased) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 6.2383
18
 
19
  ## Model description
20
 
@@ -33,35 +33,59 @@ More information needed
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
- - learning_rate: 5e-05
37
  - train_batch_size: 2
38
  - eval_batch_size: 2
39
  - seed: 250500
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
- - num_epochs: 10
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
- | 5.9983 | 0.83 | 500 | 6.2383 |
49
- | 6.2621 | 1.65 | 1000 | 6.2383 |
50
- | 6.2492 | 2.48 | 1500 | 6.2383 |
51
- | 6.2548 | 3.3 | 2000 | 6.2383 |
52
- | 6.2515 | 4.13 | 2500 | 6.2383 |
53
- | 6.2543 | 4.95 | 3000 | 6.2383 |
54
- | 6.2545 | 5.78 | 3500 | 6.2383 |
55
- | 6.247 | 6.6 | 4000 | 6.2383 |
56
- | 6.2499 | 7.43 | 4500 | 6.2383 |
57
- | 6.2434 | 8.25 | 5000 | 6.2383 |
58
- | 6.2416 | 9.08 | 5500 | 6.2383 |
59
- | 6.2425 | 9.9 | 6000 | 6.2383 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
63
 
64
- - Transformers 4.14.1
65
  - Pytorch 1.8.0+cu101
66
- - Datasets 1.16.1
67
  - Tokenizers 0.10.3
14
 
15
  This model is a fine-tuned version of [microsoft/layoutlmv2-large-uncased](https://huggingface.co/microsoft/layoutlmv2-large-uncased) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 8.5806
18
 
19
  ## Model description
20
 
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
+ - learning_rate: 2e-05
37
  - train_batch_size: 2
38
  - eval_batch_size: 2
39
  - seed: 250500
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
+ - num_epochs: 6
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
+ | No log | 0.17 | 100 | 4.6181 |
49
+ | No log | 0.33 | 200 | 4.3357 |
50
+ | No log | 0.5 | 300 | 4.3897 |
51
+ | No log | 0.66 | 400 | 4.8238 |
52
+ | 4.4277 | 0.83 | 500 | 3.9088 |
53
+ | 4.4277 | 0.99 | 600 | 3.6063 |
54
+ | 4.4277 | 1.16 | 700 | 3.4278 |
55
+ | 4.4277 | 1.32 | 800 | 3.5428 |
56
+ | 4.4277 | 1.49 | 900 | 3.4331 |
57
+ | 3.0413 | 1.65 | 1000 | 3.3699 |
58
+ | 3.0413 | 1.82 | 1100 | 3.3622 |
59
+ | 3.0413 | 1.98 | 1200 | 3.5294 |
60
+ | 3.0413 | 2.15 | 1300 | 3.7918 |
61
+ | 3.0413 | 2.31 | 1400 | 3.4007 |
62
+ | 2.0843 | 2.48 | 1500 | 4.0296 |
63
+ | 2.0843 | 2.64 | 1600 | 4.1852 |
64
+ | 2.0843 | 2.81 | 1700 | 3.6690 |
65
+ | 2.0843 | 2.97 | 1800 | 3.6089 |
66
+ | 2.0843 | 3.14 | 1900 | 5.5534 |
67
+ | 1.7527 | 3.3 | 2000 | 4.7498 |
68
+ | 1.7527 | 3.47 | 2100 | 5.2691 |
69
+ | 1.7527 | 3.63 | 2200 | 5.1324 |
70
+ | 1.7527 | 3.8 | 2300 | 4.5912 |
71
+ | 1.7527 | 3.96 | 2400 | 4.1727 |
72
+ | 1.2037 | 4.13 | 2500 | 6.1174 |
73
+ | 1.2037 | 4.29 | 2600 | 5.7172 |
74
+ | 1.2037 | 4.46 | 2700 | 5.8843 |
75
+ | 1.2037 | 4.62 | 2800 | 6.4232 |
76
+ | 1.2037 | 4.79 | 2900 | 7.4486 |
77
+ | 0.8386 | 4.95 | 3000 | 7.1946 |
78
+ | 0.8386 | 5.12 | 3100 | 7.9869 |
79
+ | 0.8386 | 5.28 | 3200 | 8.0310 |
80
+ | 0.8386 | 5.45 | 3300 | 8.2954 |
81
+ | 0.8386 | 5.61 | 3400 | 8.5361 |
82
+ | 0.4389 | 5.78 | 3500 | 8.6040 |
83
+ | 0.4389 | 5.94 | 3600 | 8.5806 |
84
 
85
 
86
  ### Framework versions
87
 
88
+ - Transformers 4.15.0
89
  - Pytorch 1.8.0+cu101
90
+ - Datasets 1.17.0
91
  - Tokenizers 0.10.3