Hugging Face's logo

rdsmaia
/
t5_small_en-pt

Text2Text Generation
Transformers
PyTorch
t5
Generated from Trainer
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
1
rdsmaia commited on
Commit
e1a70e5
·
1 Parent(s): e739f60

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +57 -22
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Bleu
24
  type: bleu
25
- value: 1.0231
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the opus_books dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 2.8376
36
- - Bleu: 1.0231
37
- - Gen Len: 18.0036
38
 
39
  ## Model description
40
 
@@ -53,33 +53,68 @@ More information needed
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
- - learning_rate: 2e-05
57
- - train_batch_size: 16
58
  - eval_batch_size: 16
59
  - seed: 42
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
- - num_epochs: 15
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
68
- | No log | 1.0 | 71 | 3.5794 | 0.7605 | 17.1246 |
69
- | No log | 2.0 | 142 | 3.3608 | 0.7738 | 17.3416 |
70
- | No log | 3.0 | 213 | 3.2292 | 1.1327 | 17.5765 |
71
- | No log | 4.0 | 284 | 3.1332 | 1.3912 | 17.7616 |
72
- | No log | 5.0 | 355 | 3.0629 | 1.2543 | 17.8363 |
73
- | No log | 6.0 | 426 | 3.0078 | 1.1736 | 17.9039 |
74
- | No log | 7.0 | 497 | 2.9674 | 1.1007 | 17.9573 |
75
- | 3.5712 | 8.0 | 568 | 2.9342 | 1.1528 | 17.9181 |
76
- | 3.5712 | 9.0 | 639 | 2.9071 | 1.1419 | 17.9324 |
77
- | 3.5712 | 10.0 | 710 | 2.8845 | 1.0593 | 17.9075 |
78
- | 3.5712 | 11.0 | 781 | 2.8676 | 1.1076 | 17.9502 |
79
- | 3.5712 | 12.0 | 852 | 2.8537 | 1.1294 | 18.0 |
80
- | 3.5712 | 13.0 | 923 | 2.8449 | 1.0701 | 17.9893 |
81
- | 3.5712 | 14.0 | 994 | 2.8396 | 1.0798 | 18.0 |
82
- | 3.1747 | 15.0 | 1065 | 2.8376 | 1.0231 | 18.0036 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
83
 
84
 
85
  ### Framework versions
 
22
  metrics:
23
  - name: Bleu
24
  type: bleu
25
+ value: 6.3078
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the opus_books dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 1.5396
36
+ - Bleu: 6.3078
37
+ - Gen Len: 17.9644
38
 
39
  ## Model description
40
 
 
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
+ - learning_rate: 0.0002
57
+ - train_batch_size: 32
58
  - eval_batch_size: 16
59
  - seed: 42
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
+ - num_epochs: 50
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
68
+ | No log | 1.0 | 36 | 2.9730 | 1.4396 | 17.9786 |
69
+ | No log | 2.0 | 72 | 2.6798 | 1.7604 | 18.1495 |
70
+ | No log | 3.0 | 108 | 2.4812 | 1.8606 | 18.2954 |
71
+ | No log | 4.0 | 144 | 2.3366 | 2.1379 | 18.2028 |
72
+ | No log | 5.0 | 180 | 2.2165 | 2.3636 | 18.1922 |
73
+ | No log | 6.0 | 216 | 2.1192 | 2.7925 | 18.1815 |
74
+ | No log | 7.0 | 252 | 2.0400 | 3.2243 | 18.1957 |
75
+ | No log | 8.0 | 288 | 1.9703 | 3.6722 | 18.1886 |
76
+ | No log | 9.0 | 324 | 1.9227 | 3.6416 | 18.1851 |
77
+ | No log | 10.0 | 360 | 1.8762 | 3.9984 | 18.1851 |
78
+ | No log | 11.0 | 396 | 1.8303 | 4.0942 | 18.1103 |
79
+ | No log | 12.0 | 432 | 1.8044 | 4.4425 | 18.1388 |
80
+ | No log | 13.0 | 468 | 1.7719 | 4.3346 | 18.1423 |
81
+ | 2.281 | 14.0 | 504 | 1.7477 | 4.6716 | 18.1032 |
82
+ | 2.281 | 15.0 | 540 | 1.7256 | 4.7874 | 18.1139 |
83
+ | 2.281 | 16.0 | 576 | 1.7057 | 4.8878 | 18.0783 |
84
+ | 2.281 | 17.0 | 612 | 1.6871 | 4.8045 | 18.0819 |
85
+ | 2.281 | 18.0 | 648 | 1.6770 | 4.9783 | 18.0676 |
86
+ | 2.281 | 19.0 | 684 | 1.6542 | 5.1069 | 18.0107 |
87
+ | 2.281 | 20.0 | 720 | 1.6414 | 4.902 | 18.0569 |
88
+ | 2.281 | 21.0 | 756 | 1.6326 | 5.0385 | 18.0214 |
89
+ | 2.281 | 22.0 | 792 | 1.6228 | 5.1533 | 18.0534 |
90
+ | 2.281 | 23.0 | 828 | 1.6233 | 5.397 | 18.0285 |
91
+ | 2.281 | 24.0 | 864 | 1.6076 | 5.4458 | 18.0214 |
92
+ | 2.281 | 25.0 | 900 | 1.5995 | 5.5752 | 18.0712 |
93
+ | 2.281 | 26.0 | 936 | 1.5938 | 5.3835 | 18.0925 |
94
+ | 2.281 | 27.0 | 972 | 1.5863 | 5.6135 | 18.0107 |
95
+ | 1.3904 | 28.0 | 1008 | 1.5780 | 5.8076 | 18.0356 |
96
+ | 1.3904 | 29.0 | 1044 | 1.5757 | 5.8528 | 18.0641 |
97
+ | 1.3904 | 30.0 | 1080 | 1.5721 | 5.8875 | 18.0285 |
98
+ | 1.3904 | 31.0 | 1116 | 1.5648 | 6.1429 | 18.0498 |
99
+ | 1.3904 | 32.0 | 1152 | 1.5596 | 6.0269 | 18.0819 |
100
+ | 1.3904 | 33.0 | 1188 | 1.5592 | 6.2233 | 18.0427 |
101
+ | 1.3904 | 34.0 | 1224 | 1.5552 | 6.0874 | 18.0569 |
102
+ | 1.3904 | 35.0 | 1260 | 1.5542 | 6.2611 | 18.0463 |
103
+ | 1.3904 | 36.0 | 1296 | 1.5493 | 6.1328 | 18.0391 |
104
+ | 1.3904 | 37.0 | 1332 | 1.5509 | 6.2341 | 18.0356 |
105
+ | 1.3904 | 38.0 | 1368 | 1.5455 | 6.2754 | 18.0036 |
106
+ | 1.3904 | 39.0 | 1404 | 1.5468 | 6.2263 | 18.0071 |
107
+ | 1.3904 | 40.0 | 1440 | 1.5446 | 6.1178 | 17.9929 |
108
+ | 1.3904 | 41.0 | 1476 | 1.5436 | 6.3536 | 17.9964 |
109
+ | 1.1159 | 42.0 | 1512 | 1.5426 | 6.296 | 17.9715 |
110
+ | 1.1159 | 43.0 | 1548 | 1.5402 | 6.1919 | 18.0356 |
111
+ | 1.1159 | 44.0 | 1584 | 1.5386 | 6.2256 | 18.0356 |
112
+ | 1.1159 | 45.0 | 1620 | 1.5392 | 6.2119 | 18.0356 |
113
+ | 1.1159 | 46.0 | 1656 | 1.5404 | 6.3696 | 18.032 |
114
+ | 1.1159 | 47.0 | 1692 | 1.5390 | 6.3779 | 17.9964 |
115
+ | 1.1159 | 48.0 | 1728 | 1.5392 | 6.2079 | 18.0107 |
116
+ | 1.1159 | 49.0 | 1764 | 1.5396 | 6.3334 | 17.968 |
117
+ | 1.1159 | 50.0 | 1800 | 1.5396 | 6.3078 | 17.9644 |
118
 
119
 
120
  ### Framework versions