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
58a077b
•
1 Parent(s): c569dfc

Rerun first training run on complete dataset

Browse files
Files changed (1) hide show
  1. README.md +22 -22
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: 4.5684
20
- - Top-5-accuracy: 0.1158
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: 0.03
40
  - train_batch_size: 16
41
  - eval_batch_size: 16
42
  - seed: 42
@@ -48,25 +48,25 @@ The following hyperparameters were used during training:
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Top-5-accuracy |
50
  |:-------------:|:------:|:----:|:---------------:|:--------------:|
51
- | 7.7153 | 0.2577 | 200 | 5.3876 | 0.0 |
52
- | 6.8602 | 0.5155 | 400 | 4.8652 | 0.0 |
53
- | 6.3689 | 0.7732 | 600 | 4.6435 | 0.0 |
54
- | 6.2303 | 1.0309 | 800 | 4.6293 | 0.0579 |
55
- | 6.0898 | 1.2887 | 1000 | 4.6395 | 0.0289 |
56
- | 6.0367 | 1.5464 | 1200 | 4.5855 | 0.0289 |
57
- | 5.8512 | 1.8041 | 1400 | 4.5860 | 0.0579 |
58
- | 5.9489 | 2.0619 | 1600 | 4.5672 | 0.0868 |
59
- | 5.7601 | 2.3196 | 1800 | 4.5522 | 0.0579 |
60
- | 5.7379 | 2.5773 | 2000 | 4.5572 | 0.0868 |
61
- | 5.7397 | 2.8351 | 2200 | 4.5559 | 0.0579 |
62
- | 5.7488 | 3.0928 | 2400 | 4.5769 | 0.1447 |
63
- | 5.7581 | 3.3505 | 2600 | 4.5421 | 0.1158 |
64
- | 5.6448 | 3.6082 | 2800 | 4.5174 | 0.1447 |
65
- | 5.6551 | 3.8660 | 3000 | 4.5773 | 0.1158 |
66
- | 5.6971 | 4.1237 | 3200 | 4.5495 | 0.0868 |
67
- | 5.7085 | 4.3814 | 3400 | 4.5392 | 0.1447 |
68
- | 5.6689 | 4.6392 | 3600 | 4.5707 | 0.1158 |
69
- | 5.5422 | 4.8969 | 3800 | 4.5684 | 0.1158 |
70
 
71
 
72
  ### 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: 5.6126
20
+ - Top-5-accuracy: 0.0
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 0.001
40
  - train_batch_size: 16
41
  - eval_batch_size: 16
42
  - seed: 42
 
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Top-5-accuracy |
50
  |:-------------:|:------:|:----:|:---------------:|:--------------:|
51
+ | 8.4166 | 0.2577 | 200 | 5.9848 | 0.0 |
52
+ | 8.297 | 0.5155 | 400 | 5.9446 | 0.0 |
53
+ | 8.0509 | 0.7732 | 600 | 5.8986 | 0.0 |
54
+ | 8.1095 | 1.0309 | 800 | 5.8153 | 0.0 |
55
+ | 7.9101 | 1.2887 | 1000 | 5.7811 | 0.0 |
56
+ | 8.0255 | 1.5464 | 1200 | 5.7496 | 0.0 |
57
+ | 8.0218 | 1.8041 | 1400 | 5.7238 | 0.0 |
58
+ | 8.0497 | 2.0619 | 1600 | 5.7016 | 0.0 |
59
+ | 8.1829 | 2.3196 | 1800 | 5.6813 | 0.0 |
60
+ | 8.0591 | 2.5773 | 2000 | 5.6719 | 0.0 |
61
+ | 8.0816 | 2.8351 | 2200 | 5.6573 | 0.0 |
62
+ | 7.9825 | 3.0928 | 2400 | 5.6475 | 0.0 |
63
+ | 8.1364 | 3.3505 | 2600 | 5.6383 | 0.0 |
64
+ | 7.9707 | 3.6082 | 2800 | 5.6298 | 0.0 |
65
+ | 7.9173 | 3.8660 | 3000 | 5.6232 | 0.0 |
66
+ | 8.0502 | 4.1237 | 3200 | 5.6226 | 0.0 |
67
+ | 8.1764 | 4.3814 | 3400 | 5.6163 | 0.0 |
68
+ | 7.9046 | 4.6392 | 3600 | 5.6141 | 0.0 |
69
+ | 7.7162 | 4.8969 | 3800 | 5.6126 | 0.0 |
70
 
71
 
72
  ### Framework versions