Hugging Face's logo

e22vvb
/
EN_mt5-base_15_spider

Text2Text Generation
Transformers
PyTorch
TensorBoard
mt5
generated_from_trainer
Inference Endpoints
text-generation-inference
Model card Files Metrics Community
e22vvb commited on
Commit
5532781
•
1 Parent(s): 9d552ed

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +22 -22
README.md CHANGED
@@ -14,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.4677
18
- - Rouge2 Precision: 0.5384
19
- - Rouge2 Recall: 0.3712
20
- - Rouge2 Fmeasure: 0.413
21
 
22
  ## Model description
23
 
@@ -37,7 +37,7 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 5e-05
40
- - train_batch_size: 16
41
  - eval_batch_size: 16
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -46,23 +46,23 @@ The following hyperparameters were used during training:
46
 
47
  ### Training results
48
 
49
- | Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
50
- |:-------------:|:-----:|:----:|:---------------:|:----------------:|:-------------:|:---------------:|
51
- | No log | 1.0 | 438 | 1.7721 | 0.0 | 0.0 | 0.0 |
52
- | 10.9203 | 2.0 | 876 | 1.5463 | 0.0048 | 0.0008 | 0.0015 |
53
- | 1.7683 | 3.0 | 1314 | 1.3623 | 0.0164 | 0.0008 | 0.0014 |
54
- | 1.6091 | 4.0 | 1752 | 1.0649 | 0.0036 | 0.0012 | 0.0014 |
55
- | 1.2253 | 5.0 | 2190 | 0.3614 | 0.2985 | 0.1906 | 0.2088 |
56
- | 0.3973 | 6.0 | 2628 | 0.3133 | 0.4588 | 0.2914 | 0.3317 |
57
- | 0.2032 | 7.0 | 3066 | 0.3007 | 0.5065 | 0.3375 | 0.3799 |
58
- | 0.1597 | 8.0 | 3504 | 0.3109 | 0.5212 | 0.3579 | 0.3988 |
59
- | 0.1597 | 9.0 | 3942 | 0.3431 | 0.5194 | 0.3529 | 0.3931 |
60
- | 0.1385 | 10.0 | 4380 | 0.4681 | 0.5315 | 0.3636 | 0.4051 |
61
- | 0.1285 | 11.0 | 4818 | 0.5505 | 0.5331 | 0.3666 | 0.4074 |
62
- | 0.1228 | 12.0 | 5256 | 0.5351 | 0.5331 | 0.3645 | 0.4063 |
63
- | 0.1092 | 13.0 | 5694 | 0.4808 | 0.5355 | 0.3697 | 0.4102 |
64
- | 0.1078 | 14.0 | 6132 | 0.4519 | 0.5407 | 0.3727 | 0.4144 |
65
- | 0.1034 | 15.0 | 6570 | 0.4677 | 0.5384 | 0.3712 | 0.413 |
66
 
67
 
68
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.3557
18
+ - Rouge2 Precision: 0.6104
19
+ - Rouge2 Recall: 0.4164
20
+ - Rouge2 Fmeasure: 0.4669
21
 
22
  ## Model description
23
 
 
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 5e-05
40
+ - train_batch_size: 12
41
  - eval_batch_size: 16
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 
46
 
47
  ### Training results
48
 
49
+ | Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
50
+ |:-------------:|:-----:|:-----:|:---------------:|:----------------:|:-------------:|:---------------:|
51
+ | 9.1945 | 1.0 | 722 | 1.9988 | 0.2524 | 0.1722 | 0.1809 |
52
+ | 2.7281 | 2.0 | 1444 | 0.3721 | 0.4727 | 0.3003 | 0.3376 |
53
+ | 0.2067 | 3.0 | 2166 | 0.3160 | 0.5435 | 0.3541 | 0.4033 |
54
+ | 0.1692 | 4.0 | 2888 | 0.3153 | 0.5628 | 0.3863 | 0.4314 |
55
+ | 0.1252 | 5.0 | 3610 | 0.3225 | 0.5764 | 0.3907 | 0.4381 |
56
+ | 0.1117 | 6.0 | 4332 | 0.3165 | 0.5707 | 0.3891 | 0.435 |
57
+ | 0.0973 | 7.0 | 5054 | 0.3269 | 0.5693 | 0.3868 | 0.4335 |
58
+ | 0.0897 | 8.0 | 5776 | 0.3365 | 0.585 | 0.4009 | 0.4478 |
59
+ | 0.082 | 9.0 | 6498 | 0.3407 | 0.6035 | 0.4129 | 0.4619 |
60
+ | 0.0721 | 10.0 | 7220 | 0.3453 | 0.5963 | 0.4089 | 0.4568 |
61
+ | 0.0703 | 11.0 | 7942 | 0.3434 | 0.6032 | 0.4137 | 0.4629 |
62
+ | 0.0638 | 12.0 | 8664 | 0.3505 | 0.6077 | 0.4178 | 0.4667 |
63
+ | 0.0615 | 13.0 | 9386 | 0.3509 | 0.611 | 0.4169 | 0.467 |
64
+ | 0.0599 | 14.0 | 10108 | 0.3554 | 0.608 | 0.4153 | 0.4651 |
65
+ | 0.058 | 15.0 | 10830 | 0.3557 | 0.6104 | 0.4164 | 0.4669 |
66
 
67
 
68
  ### Framework versions