Hugging Face's logo

zera09
/
long_t5

TensorBoard
Safetensors
longt5
Generated from Trainer
Model card Files Metrics Community
zera09 commited on
Commit
3202021
·
verified ·
1 Parent(s): cfdaf66

End of training

Browse files
Files changed (1) hide show
  1. README.md +29 -14
README.md CHANGED
@@ -17,12 +17,12 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.5355
21
- - Rouge1: 0.4922
22
- - Rouge2: 0.3075
23
- - Rougel: 0.448
24
- - Rougelsum: 0.4476
25
- - Gen Len: 25.2419
26
 
27
  ## Model description
28
 
@@ -42,22 +42,37 @@ More information needed
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 2e-05
45
- - train_batch_size: 2
46
- - eval_batch_size: 2
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
- - num_epochs: 5
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
55
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
56
- | 2.16 | 1.0 | 3200 | 1.6302 | 0.4745 | 0.2896 | 0.4315 | 0.4311 | 25.1044 |
57
- | 1.9936 | 2.0 | 6400 | 1.5803 | 0.4888 | 0.3046 | 0.4468 | 0.4463 | 25.2094 |
58
- | 1.8414 | 3.0 | 9600 | 1.5484 | 0.4905 | 0.3048 | 0.446 | 0.4455 | 25.695 |
59
- | 1.8103 | 4.0 | 12800 | 1.5363 | 0.4903 | 0.3058 | 0.4464 | 0.4465 | 24.9725 |
60
- | 1.7387 | 5.0 | 16000 | 1.5355 | 0.4922 | 0.3075 | 0.448 | 0.4476 | 25.2419 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
 
62
 
63
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.4850
21
+ - Rouge1: 0.5287
22
+ - Rouge2: 0.3416
23
+ - Rougel: 0.4817
24
+ - Rougelsum: 0.4819
25
+ - Gen Len: 25.7612
26
 
27
  ## Model description
28
 
 
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 2e-05
45
+ - train_batch_size: 4
46
+ - eval_batch_size: 4
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
+ - num_epochs: 20
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
55
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
56
+ | 2.2416 | 1.0 | 1600 | 1.6572 | 0.4727 | 0.2853 | 0.4283 | 0.4282 | 26.6931 |
57
+ | 2.0392 | 2.0 | 3200 | 1.5864 | 0.4876 | 0.3019 | 0.4422 | 0.4423 | 25.9194 |
58
+ | 1.9573 | 3.0 | 4800 | 1.5519 | 0.4968 | 0.3089 | 0.4509 | 0.451 | 26.8606 |
59
+ | 1.8886 | 4.0 | 6400 | 1.5336 | 0.4986 | 0.3111 | 0.4511 | 0.4511 | 26.2788 |
60
+ | 1.7906 | 5.0 | 8000 | 1.5149 | 0.5038 | 0.3166 | 0.4567 | 0.4567 | 25.8431 |
61
+ | 1.729 | 6.0 | 9600 | 1.5076 | 0.505 | 0.3194 | 0.4584 | 0.4586 | 25.8794 |
62
+ | 1.6885 | 7.0 | 11200 | 1.4985 | 0.5119 | 0.3264 | 0.465 | 0.4651 | 25.86 |
63
+ | 1.6507 | 8.0 | 12800 | 1.4968 | 0.5154 | 0.329 | 0.4678 | 0.4678 | 26.0975 |
64
+ | 1.6412 | 9.0 | 14400 | 1.4854 | 0.515 | 0.3289 | 0.4687 | 0.4689 | 25.4875 |
65
+ | 1.5875 | 10.0 | 16000 | 1.4863 | 0.5178 | 0.3327 | 0.4715 | 0.4719 | 25.745 |
66
+ | 1.5151 | 11.0 | 17600 | 1.4864 | 0.5197 | 0.3357 | 0.4744 | 0.4744 | 25.5288 |
67
+ | 1.509 | 12.0 | 19200 | 1.4819 | 0.5213 | 0.3369 | 0.4755 | 0.4755 | 25.9488 |
68
+ | 1.5232 | 13.0 | 20800 | 1.4833 | 0.5237 | 0.3377 | 0.4773 | 0.4778 | 25.5394 |
69
+ | 1.4704 | 14.0 | 22400 | 1.4840 | 0.525 | 0.34 | 0.4791 | 0.4792 | 25.46 |
70
+ | 1.4585 | 15.0 | 24000 | 1.4836 | 0.5239 | 0.3384 | 0.4779 | 0.4781 | 26.0238 |
71
+ | 1.4607 | 16.0 | 25600 | 1.4848 | 0.5265 | 0.3396 | 0.4792 | 0.4794 | 25.8919 |
72
+ | 1.42 | 17.0 | 27200 | 1.4889 | 0.5274 | 0.3409 | 0.4803 | 0.4804 | 26.2381 |
73
+ | 1.4461 | 18.0 | 28800 | 1.4855 | 0.5283 | 0.342 | 0.4814 | 0.4815 | 25.605 |
74
+ | 1.4384 | 19.0 | 30400 | 1.4852 | 0.5284 | 0.3412 | 0.4813 | 0.4816 | 25.8419 |
75
+ | 1.4412 | 20.0 | 32000 | 1.4850 | 0.5287 | 0.3416 | 0.4817 | 0.4819 | 25.7612 |
76
 
77
 
78
  ### Framework versions