Hugging Face's logo

Anise1006
/
layoutlmv2-base-uncased_finetuned_docvqa

Document Question Answering
Transformers
PyTorch
TensorBoard
layoutlmv2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Anise1006 commited on
Commit
98ae7ff
·
1 Parent(s): 16b22d6

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +100 -2
README.md CHANGED
@@ -12,6 +12,8 @@ should probably proofread and complete it, then remove this comment. -->
12
  # layoutlmv2-base-uncased_finetuned_docvqa
13
 
14
  This model was trained from scratch on the None dataset.
 
 
15
 
16
  ## Model description
17
 
@@ -31,13 +33,109 @@ More information needed
31
 
32
  The following hyperparameters were used during training:
33
  - learning_rate: 5e-05
34
- - train_batch_size: 8
35
- - eval_batch_size: 16
36
  - seed: 42
37
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
38
  - lr_scheduler_type: linear
39
  - num_epochs: 20
40
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41
  ### Framework versions
42
 
43
  - Transformers 4.30.2
 
12
  # layoutlmv2-base-uncased_finetuned_docvqa
13
 
14
  This model was trained from scratch on the None dataset.
15
+ It achieves the following results on the evaluation set:
16
+ - Loss: 4.6973
17
 
18
  ## Model description
19
 
 
33
 
34
  The following hyperparameters were used during training:
35
  - learning_rate: 5e-05
36
+ - train_batch_size: 4
37
+ - eval_batch_size: 8
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
  - num_epochs: 20
42
 
43
+ ### Training results
44
+
45
+ | Training Loss | Epoch | Step | Validation Loss |
46
+ |:-------------:|:-----:|:----:|:---------------:|
47
+ | 5.3322 | 0.22 | 50 | 4.6214 |
48
+ | 4.4006 | 0.44 | 100 | 4.1705 |
49
+ | 3.9967 | 0.66 | 150 | 3.8115 |
50
+ | 3.8349 | 0.88 | 200 | 3.5599 |
51
+ | 3.4185 | 1.11 | 250 | 3.2811 |
52
+ | 3.1129 | 1.33 | 300 | 3.3014 |
53
+ | 2.9186 | 1.55 | 350 | 3.2583 |
54
+ | 2.8937 | 1.77 | 400 | 2.6806 |
55
+ | 2.5872 | 1.99 | 450 | 2.4617 |
56
+ | 1.934 | 2.21 | 500 | 2.4098 |
57
+ | 1.9086 | 2.43 | 550 | 2.3926 |
58
+ | 1.7456 | 2.65 | 600 | 2.4048 |
59
+ | 1.6092 | 2.88 | 650 | 2.2064 |
60
+ | 1.5114 | 3.1 | 700 | 2.3882 |
61
+ | 1.3176 | 3.32 | 750 | 2.2211 |
62
+ | 1.2236 | 3.54 | 800 | 3.5792 |
63
+ | 1.4123 | 3.76 | 850 | 2.0391 |
64
+ | 1.2434 | 3.98 | 900 | 2.3563 |
65
+ | 0.8727 | 4.2 | 950 | 2.6344 |
66
+ | 0.9026 | 4.42 | 1000 | 3.1071 |
67
+ | 1.0698 | 4.65 | 1050 | 3.1795 |
68
+ | 0.9226 | 4.87 | 1100 | 3.0897 |
69
+ | 1.0639 | 5.09 | 1150 | 3.2822 |
70
+ | 0.7568 | 5.31 | 1200 | 3.0442 |
71
+ | 0.8192 | 5.53 | 1250 | 2.8516 |
72
+ | 0.5137 | 5.75 | 1300 | 3.1520 |
73
+ | 0.7053 | 5.97 | 1350 | 2.7512 |
74
+ | 0.4587 | 6.19 | 1400 | 2.8772 |
75
+ | 0.4965 | 6.42 | 1450 | 3.1186 |
76
+ | 0.4079 | 6.64 | 1500 | 3.0887 |
77
+ | 0.6135 | 6.86 | 1550 | 3.0537 |
78
+ | 0.7518 | 7.08 | 1600 | 3.2562 |
79
+ | 0.336 | 7.3 | 1650 | 3.3709 |
80
+ | 0.618 | 7.52 | 1700 | 2.5948 |
81
+ | 0.3473 | 7.74 | 1750 | 3.1784 |
82
+ | 0.3811 | 7.96 | 1800 | 3.3420 |
83
+ | 0.2749 | 8.19 | 1850 | 3.5958 |
84
+ | 0.2916 | 8.41 | 1900 | 3.7811 |
85
+ | 0.2634 | 8.63 | 1950 | 3.8935 |
86
+ | 0.4594 | 8.85 | 2000 | 3.4889 |
87
+ | 0.2957 | 9.07 | 2050 | 3.5717 |
88
+ | 0.1503 | 9.29 | 2100 | 3.8861 |
89
+ | 0.2986 | 9.51 | 2150 | 3.8007 |
90
+ | 0.3732 | 9.73 | 2200 | 3.5632 |
91
+ | 0.282 | 9.96 | 2250 | 3.1815 |
92
+ | 0.1781 | 10.18 | 2300 | 3.8386 |
93
+ | 0.1678 | 10.4 | 2350 | 3.9624 |
94
+ | 0.1813 | 10.62 | 2400 | 4.1113 |
95
+ | 0.3601 | 10.84 | 2450 | 3.9760 |
96
+ | 0.2374 | 11.06 | 2500 | 3.8321 |
97
+ | 0.1251 | 11.28 | 2550 | 4.0945 |
98
+ | 0.0741 | 11.5 | 2600 | 4.1810 |
99
+ | 0.2846 | 11.73 | 2650 | 3.9240 |
100
+ | 0.1964 | 11.95 | 2700 | 3.7833 |
101
+ | 0.1413 | 12.17 | 2750 | 3.9830 |
102
+ | 0.0696 | 12.39 | 2800 | 4.4270 |
103
+ | 0.1401 | 12.61 | 2850 | 4.6113 |
104
+ | 0.1267 | 12.83 | 2900 | 4.3713 |
105
+ | 0.0689 | 13.05 | 2950 | 4.4582 |
106
+ | 0.1636 | 13.27 | 3000 | 4.6710 |
107
+ | 0.1935 | 13.5 | 3050 | 4.2833 |
108
+ | 0.0936 | 13.72 | 3100 | 4.5357 |
109
+ | 0.1001 | 13.94 | 3150 | 4.2660 |
110
+ | 0.0509 | 14.16 | 3200 | 4.3863 |
111
+ | 0.0193 | 14.38 | 3250 | 4.5554 |
112
+ | 0.031 | 14.6 | 3300 | 4.4455 |
113
+ | 0.0163 | 14.82 | 3350 | 4.4985 |
114
+ | 0.1558 | 15.04 | 3400 | 4.5935 |
115
+ | 0.0707 | 15.27 | 3450 | 4.4791 |
116
+ | 0.0089 | 15.49 | 3500 | 4.6358 |
117
+ | 0.0881 | 15.71 | 3550 | 4.5644 |
118
+ | 0.1348 | 15.93 | 3600 | 4.5336 |
119
+ | 0.0176 | 16.15 | 3650 | 4.5465 |
120
+ | 0.0679 | 16.37 | 3700 | 4.4872 |
121
+ | 0.0138 | 16.59 | 3750 | 4.3819 |
122
+ | 0.0211 | 16.81 | 3800 | 4.5024 |
123
+ | 0.0506 | 17.04 | 3850 | 4.4761 |
124
+ | 0.0051 | 17.26 | 3900 | 4.5898 |
125
+ | 0.075 | 17.48 | 3950 | 4.5427 |
126
+ | 0.0089 | 17.7 | 4000 | 4.6244 |
127
+ | 0.0298 | 17.92 | 4050 | 4.6440 |
128
+ | 0.0034 | 18.14 | 4100 | 4.6486 |
129
+ | 0.0594 | 18.36 | 4150 | 4.6438 |
130
+ | 0.0179 | 18.58 | 4200 | 4.6552 |
131
+ | 0.0081 | 18.81 | 4250 | 4.6508 |
132
+ | 0.0046 | 19.03 | 4300 | 4.6808 |
133
+ | 0.0618 | 19.25 | 4350 | 4.6856 |
134
+ | 0.0033 | 19.47 | 4400 | 4.6763 |
135
+ | 0.0062 | 19.69 | 4450 | 4.6892 |
136
+ | 0.0404 | 19.91 | 4500 | 4.6973 |
137
+
138
+
139
  ### Framework versions
140
 
141
  - Transformers 4.30.2