Hugging Face's logo

esahit
/
ul2-large-dutch-finetuned-oba-book-search

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
esahit commited on
Commit
8254765
·
verified ·
1 Parent(s): fc87912

Second training rerun on increased dataset

Browse files
Files changed (1) hide show
  1. README.md +73 -73
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [yhavinga/ul2-large-dutch](https://huggingface.co/yhavinga/ul2-large-dutch) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 3.9161
20
- - Top-5-accuracy: 4.3582
21
 
22
  ## Model description
23
 
@@ -36,7 +36,7 @@ More information needed
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
- - learning_rate: 0.3
40
  - train_batch_size: 16
41
  - eval_batch_size: 16
42
  - seed: 42
@@ -49,76 +49,76 @@ The following hyperparameters were used during training:
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Top-5-accuracy |
51
  |:-------------:|:------:|:-----:|:---------------:|:--------------:|
52
- | 6.7818 | 0.0424 | 500 | 4.6703 | 0.0995 |
53
- | 6.1358 | 0.0848 | 1000 | 4.6934 | 0.0 |
54
- | 5.7889 | 0.1272 | 1500 | 4.4938 | 0.0995 |
55
- | 5.5769 | 0.1696 | 2000 | 4.4436 | 0.1592 |
56
- | 5.5432 | 0.2120 | 2500 | 4.5134 | 0.1393 |
57
- | 5.4227 | 0.2544 | 3000 | 4.3984 | 0.3184 |
58
- | 5.462 | 0.2968 | 3500 | 4.3750 | 0.3582 |
59
- | 5.3557 | 0.3392 | 4000 | 4.3295 | 0.7363 |
60
- | 5.3069 | 0.3815 | 4500 | 4.3453 | 0.5771 |
61
- | 5.1367 | 0.4239 | 5000 | 4.2607 | 1.3930 |
62
- | 5.2098 | 0.4663 | 5500 | 4.2917 | 1.2935 |
63
- | 5.2545 | 0.5087 | 6000 | 4.2551 | 2.2687 |
64
- | 5.161 | 0.5511 | 6500 | 4.2483 | 2.0896 |
65
- | 5.0716 | 0.5935 | 7000 | 4.2450 | 2.6866 |
66
- | 5.0687 | 0.6359 | 7500 | 4.2535 | 2.4080 |
67
- | 5.0434 | 0.6783 | 8000 | 4.1765 | 3.1443 |
68
- | 5.0132 | 0.7207 | 8500 | 4.1843 | 3.2836 |
69
- | 4.9609 | 0.7631 | 9000 | 4.1817 | 3.3632 |
70
- | 4.983 | 0.8055 | 9500 | 4.1351 | 3.9403 |
71
- | 4.9225 | 0.8479 | 10000 | 4.0950 | 4.1194 |
72
- | 4.8771 | 0.8903 | 10500 | 4.1032 | 4.1194 |
73
- | 4.8627 | 0.9327 | 11000 | 4.1021 | 3.9204 |
74
- | 4.8935 | 0.9751 | 11500 | 4.0641 | 4.1791 |
75
- | 4.8718 | 1.0175 | 12000 | 4.0695 | 4.1592 |
76
- | 4.8557 | 1.0599 | 12500 | 4.0757 | 4.0597 |
77
- | 4.8136 | 1.1023 | 13000 | 4.0492 | 4.1990 |
78
- | 4.8281 | 1.1446 | 13500 | 4.0538 | 4.0 |
79
- | 4.8293 | 1.1870 | 14000 | 4.0315 | 4.3184 |
80
- | 4.831 | 1.2294 | 14500 | 4.0417 | 4.1194 |
81
- | 4.8232 | 1.2718 | 15000 | 4.0157 | 4.3383 |
82
- | 4.7911 | 1.3142 | 15500 | 4.0246 | 4.3383 |
83
- | 4.7865 | 1.3566 | 16000 | 3.9911 | 4.4975 |
84
- | 4.8019 | 1.3990 | 16500 | 4.0177 | 4.2786 |
85
- | 4.796 | 1.4414 | 17000 | 4.0278 | 4.3582 |
86
- | 4.8138 | 1.4838 | 17500 | 3.9919 | 4.2587 |
87
- | 4.7367 | 1.5262 | 18000 | 3.9809 | 4.4378 |
88
- | 4.757 | 1.5686 | 18500 | 3.9729 | 4.4179 |
89
- | 4.7352 | 1.6110 | 19000 | 3.9750 | 4.3980 |
90
- | 4.7663 | 1.6534 | 19500 | 3.9824 | 4.3184 |
91
- | 4.6772 | 1.6958 | 20000 | 3.9843 | 4.3383 |
92
- | 4.7573 | 1.7382 | 20500 | 3.9641 | 4.2189 |
93
- | 4.7402 | 1.7806 | 21000 | 3.9654 | 4.3980 |
94
- | 4.7006 | 1.8230 | 21500 | 3.9557 | 4.2587 |
95
- | 4.7047 | 1.8654 | 22000 | 3.9606 | 4.2985 |
96
- | 4.6683 | 1.9077 | 22500 | 3.9558 | 4.2985 |
97
- | 4.725 | 1.9501 | 23000 | 3.9382 | 4.3383 |
98
- | 4.7176 | 1.9925 | 23500 | 3.9422 | 4.4776 |
99
- | 4.7194 | 2.0349 | 24000 | 3.9445 | 4.2985 |
100
- | 4.6886 | 2.0773 | 24500 | 3.9368 | 4.4378 |
101
- | 4.6876 | 2.1197 | 25000 | 3.9245 | 4.4179 |
102
- | 4.6877 | 2.1621 | 25500 | 3.9326 | 4.3383 |
103
- | 4.7219 | 2.2045 | 26000 | 3.9296 | 4.3781 |
104
- | 4.6815 | 2.2469 | 26500 | 3.9279 | 4.3184 |
105
- | 4.6839 | 2.2893 | 27000 | 3.9276 | 4.2587 |
106
- | 4.6103 | 2.3317 | 27500 | 3.9251 | 4.2985 |
107
- | 4.6566 | 2.3741 | 28000 | 3.9307 | 4.2985 |
108
- | 4.6523 | 2.4165 | 28500 | 3.9236 | 4.2587 |
109
- | 4.6363 | 2.4589 | 29000 | 3.9193 | 4.2786 |
110
- | 4.6575 | 2.5013 | 29500 | 3.9185 | 4.2388 |
111
- | 4.6161 | 2.5437 | 30000 | 3.9227 | 4.3184 |
112
- | 4.644 | 2.5861 | 30500 | 3.9162 | 4.2985 |
113
- | 4.6537 | 2.6285 | 31000 | 3.9169 | 4.3781 |
114
- | 4.6405 | 2.6708 | 31500 | 3.9214 | 4.3781 |
115
- | 4.7401 | 2.7132 | 32000 | 3.9191 | 4.3781 |
116
- | 4.6907 | 2.7556 | 32500 | 3.9161 | 4.3582 |
117
- | 4.7139 | 2.7980 | 33000 | 3.9169 | 4.3781 |
118
- | 4.6537 | 2.8404 | 33500 | 3.9171 | 4.3582 |
119
- | 4.7395 | 2.8828 | 34000 | 3.9162 | 4.3383 |
120
- | 4.58 | 2.9252 | 34500 | 3.9162 | 4.3582 |
121
- | 4.6671 | 2.9676 | 35000 | 3.9161 | 4.3582 |
122
 
123
 
124
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [yhavinga/ul2-large-dutch](https://huggingface.co/yhavinga/ul2-large-dutch) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 3.8688
20
+ - Top-5-accuracy: 4.1194
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 0.6
40
  - train_batch_size: 16
41
  - eval_batch_size: 16
42
  - seed: 42
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Top-5-accuracy |
51
  |:-------------:|:------:|:-----:|:---------------:|:--------------:|
52
+ | 6.4431 | 0.0424 | 500 | 4.7239 | 0.0796 |
53
+ | 6.4068 | 0.0848 | 1000 | 5.1338 | 0.0398 |
54
+ | 5.7971 | 0.1272 | 1500 | 4.6127 | 0.0199 |
55
+ | 5.452 | 0.1696 | 2000 | 4.5181 | 0.1194 |
56
+ | 5.3971 | 0.2120 | 2500 | 4.5498 | 0.1393 |
57
+ | 5.2693 | 0.2544 | 3000 | 4.3622 | 0.1393 |
58
+ | 5.2788 | 0.2968 | 3500 | 4.3456 | 0.1990 |
59
+ | 5.2129 | 0.3392 | 4000 | 4.3400 | 0.2388 |
60
+ | 5.133 | 0.3815 | 4500 | 4.3021 | 0.2786 |
61
+ | 5.0346 | 0.4239 | 5000 | 4.2458 | 0.9751 |
62
+ | 5.113 | 0.4663 | 5500 | 4.2746 | 0.7363 |
63
+ | 5.1276 | 0.5087 | 6000 | 4.2369 | 0.9552 |
64
+ | 5.0586 | 0.5511 | 6500 | 4.1962 | 1.8706 |
65
+ | 4.9369 | 0.5935 | 7000 | 4.1843 | 2.9254 |
66
+ | 4.9152 | 0.6359 | 7500 | 4.1641 | 3.0846 |
67
+ | 4.9369 | 0.6783 | 8000 | 4.1089 | 3.7413 |
68
+ | 4.9185 | 0.7207 | 8500 | 4.1150 | 3.6418 |
69
+ | 4.8469 | 0.7631 | 9000 | 4.0996 | 3.6418 |
70
+ | 4.8854 | 0.8055 | 9500 | 4.0817 | 3.5821 |
71
+ | 4.8362 | 0.8479 | 10000 | 4.0456 | 4.2587 |
72
+ | 4.7867 | 0.8903 | 10500 | 4.0699 | 3.9204 |
73
+ | 4.7926 | 0.9327 | 11000 | 4.0692 | 3.3831 |
74
+ | 4.7933 | 0.9751 | 11500 | 4.0356 | 3.1642 |
75
+ | 4.793 | 1.0175 | 12000 | 4.0607 | 2.6667 |
76
+ | 4.7664 | 1.0599 | 12500 | 4.0430 | 3.5622 |
77
+ | 4.7409 | 1.1023 | 13000 | 4.0239 | 3.8806 |
78
+ | 4.7558 | 1.1446 | 13500 | 4.0134 | 3.7413 |
79
+ | 4.7642 | 1.1870 | 14000 | 3.9884 | 3.9403 |
80
+ | 4.7298 | 1.2294 | 14500 | 4.0087 | 3.6219 |
81
+ | 4.7433 | 1.2718 | 15000 | 3.9809 | 4.0995 |
82
+ | 4.6858 | 1.3142 | 15500 | 3.9984 | 4.2985 |
83
+ | 4.7023 | 1.3566 | 16000 | 3.9655 | 4.0199 |
84
+ | 4.6963 | 1.3990 | 16500 | 3.9798 | 4.1791 |
85
+ | 4.7239 | 1.4414 | 17000 | 4.0001 | 4.0597 |
86
+ | 4.7312 | 1.4838 | 17500 | 3.9532 | 4.0796 |
87
+ | 4.6408 | 1.5262 | 18000 | 3.9487 | 4.2388 |
88
+ | 4.669 | 1.5686 | 18500 | 3.9303 | 4.1990 |
89
+ | 4.6589 | 1.6110 | 19000 | 3.9346 | 4.1393 |
90
+ | 4.6887 | 1.6534 | 19500 | 3.9563 | 3.9403 |
91
+ | 4.5856 | 1.6958 | 20000 | 3.9374 | 4.2786 |
92
+ | 4.6744 | 1.7382 | 20500 | 3.9157 | 4.0995 |
93
+ | 4.6395 | 1.7806 | 21000 | 3.9279 | 4.1393 |
94
+ | 4.6191 | 1.8230 | 21500 | 3.9259 | 3.8408 |
95
+ | 4.6256 | 1.8654 | 22000 | 3.9215 | 3.9005 |
96
+ | 4.5945 | 1.9077 | 22500 | 3.9214 | 4.0796 |
97
+ | 4.6325 | 1.9501 | 23000 | 3.9076 | 3.8607 |
98
+ | 4.6476 | 1.9925 | 23500 | 3.8955 | 4.0199 |
99
+ | 4.6362 | 2.0349 | 24000 | 3.8923 | 4.0398 |
100
+ | 4.5991 | 2.0773 | 24500 | 3.8923 | 4.3383 |
101
+ | 4.6189 | 2.1197 | 25000 | 3.8800 | 4.0 |
102
+ | 4.5933 | 2.1621 | 25500 | 3.8869 | 3.8806 |
103
+ | 4.6165 | 2.2045 | 26000 | 3.8918 | 4.0398 |
104
+ | 4.5998 | 2.2469 | 26500 | 3.8819 | 3.9602 |
105
+ | 4.5827 | 2.2893 | 27000 | 3.8848 | 3.9204 |
106
+ | 4.528 | 2.3317 | 27500 | 3.8847 | 3.9005 |
107
+ | 4.5685 | 2.3741 | 28000 | 3.8879 | 3.9204 |
108
+ | 4.5698 | 2.4165 | 28500 | 3.8739 | 3.9801 |
109
+ | 4.5472 | 2.4589 | 29000 | 3.8761 | 4.0398 |
110
+ | 4.5605 | 2.5013 | 29500 | 3.8753 | 4.0398 |
111
+ | 4.5329 | 2.5437 | 30000 | 3.8791 | 4.0796 |
112
+ | 4.5687 | 2.5861 | 30500 | 3.8698 | 4.0 |
113
+ | 4.5716 | 2.6285 | 31000 | 3.8659 | 4.0995 |
114
+ | 4.547 | 2.6708 | 31500 | 3.8713 | 4.0597 |
115
+ | 4.6466 | 2.7132 | 32000 | 3.8729 | 4.0995 |
116
+ | 4.5963 | 2.7556 | 32500 | 3.8698 | 4.1194 |
117
+ | 4.629 | 2.7980 | 33000 | 3.8703 | 4.1194 |
118
+ | 4.5859 | 2.8404 | 33500 | 3.8699 | 4.1194 |
119
+ | 4.6239 | 2.8828 | 34000 | 3.8688 | 4.1393 |
120
+ | 4.5052 | 2.9252 | 34500 | 3.8688 | 4.1393 |
121
+ | 4.5933 | 2.9676 | 35000 | 3.8688 | 4.1194 |
122
 
123
 
124
  ### Framework versions