Hugging Face's logo

din0s
/
t5-base-finetuned-en-to-it

Transformers
PyTorch
t5
text2text-generation
Generated from Trainer
Eval Results
text-generation-inference
Model card Files
xet
Community
1
din0s commited on
Commit
0c54936
·
1 Parent(s): 4870132

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +115 -0
README.md ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - ccmatrix
7
+ metrics:
8
+ - bleu
9
+ model-index:
10
+ - name: t5-base-finetuned-en-to-it
11
+ results:
12
+ - task:
13
+ name: Sequence-to-sequence Language Modeling
14
+ type: text2text-generation
15
+ dataset:
16
+ name: ccmatrix
17
+ type: ccmatrix
18
+ config: en-it
19
+ split: train[3000:12000]
20
+ args: en-it
21
+ metrics:
22
+ - name: Bleu
23
+ type: bleu
24
+ value: 20.1194
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # t5-base-finetuned-en-to-it
31
+
32
+ This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the ccmatrix dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 1.4830
35
+ - Bleu: 20.1194
36
+ - Gen Len: 51.456
37
+
38
+ ## Model description
39
+
40
+ More information needed
41
+
42
+ ## Intended uses & limitations
43
+
44
+ More information needed
45
+
46
+ ## Training and evaluation data
47
+
48
+ More information needed
49
+
50
+ ## Training procedure
51
+
52
+ ### Training hyperparameters
53
+
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 2e-05
56
+ - train_batch_size: 32
57
+ - eval_batch_size: 32
58
+ - seed: 42
59
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
+ - lr_scheduler_type: linear
61
+ - num_epochs: 40
62
+ - mixed_precision_training: Native AMP
63
+
64
+ ### Training results
65
+
66
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
67
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
68
+ | No log | 1.0 | 282 | 2.0137 | 6.5621 | 69.0227 |
69
+ | 2.4006 | 2.0 | 564 | 1.9278 | 7.2684 | 70.0333 |
70
+ | 2.4006 | 3.0 | 846 | 1.8712 | 8.6643 | 64.654 |
71
+ | 2.1423 | 4.0 | 1128 | 1.8223 | 9.3778 | 63.4453 |
72
+ | 2.1423 | 5.0 | 1410 | 1.7836 | 10.0151 | 63.778 |
73
+ | 2.0248 | 6.0 | 1692 | 1.7515 | 10.9865 | 62.224 |
74
+ | 2.0248 | 7.0 | 1974 | 1.7208 | 11.5089 | 61.2 |
75
+ | 1.9316 | 8.0 | 2256 | 1.6936 | 12.3755 | 60.1047 |
76
+ | 1.8584 | 9.0 | 2538 | 1.6731 | 12.8765 | 59.4427 |
77
+ | 1.8584 | 10.0 | 2820 | 1.6535 | 13.7278 | 57.6253 |
78
+ | 1.7949 | 11.0 | 3102 | 1.6360 | 14.2498 | 56.3913 |
79
+ | 1.7949 | 12.0 | 3384 | 1.6222 | 14.8795 | 55.346 |
80
+ | 1.7461 | 13.0 | 3666 | 1.6064 | 15.017 | 55.7473 |
81
+ | 1.7461 | 14.0 | 3948 | 1.5926 | 15.3093 | 56.0067 |
82
+ | 1.6998 | 15.0 | 4230 | 1.5803 | 15.6934 | 55.366 |
83
+ | 1.6635 | 16.0 | 4512 | 1.5707 | 16.3604 | 54.5413 |
84
+ | 1.6635 | 17.0 | 4794 | 1.5633 | 16.8086 | 53.824 |
85
+ | 1.621 | 18.0 | 5076 | 1.5515 | 17.1319 | 53.5927 |
86
+ | 1.621 | 19.0 | 5358 | 1.5450 | 17.5039 | 53.5167 |
87
+ | 1.6008 | 20.0 | 5640 | 1.5389 | 17.8012 | 53.6527 |
88
+ | 1.6008 | 21.0 | 5922 | 1.5314 | 17.7305 | 53.342 |
89
+ | 1.5656 | 22.0 | 6204 | 1.5259 | 18.1609 | 53.4033 |
90
+ | 1.5656 | 23.0 | 6486 | 1.5200 | 18.6506 | 52.226 |
91
+ | 1.5466 | 24.0 | 6768 | 1.5185 | 18.9433 | 52.2173 |
92
+ | 1.53 | 25.0 | 7050 | 1.5120 | 19.0978 | 52.022 |
93
+ | 1.53 | 26.0 | 7332 | 1.5083 | 19.1326 | 52.0527 |
94
+ | 1.5072 | 27.0 | 7614 | 1.5044 | 19.0854 | 52.2447 |
95
+ | 1.5072 | 28.0 | 7896 | 1.5002 | 19.372 | 51.7687 |
96
+ | 1.4926 | 29.0 | 8178 | 1.4977 | 19.5798 | 52.0327 |
97
+ | 1.4926 | 30.0 | 8460 | 1.4941 | 19.5161 | 51.9893 |
98
+ | 1.478 | 31.0 | 8742 | 1.4911 | 19.7821 | 51.534 |
99
+ | 1.47 | 32.0 | 9024 | 1.4897 | 19.7207 | 51.4787 |
100
+ | 1.47 | 33.0 | 9306 | 1.4888 | 19.8066 | 51.5407 |
101
+ | 1.4603 | 34.0 | 9588 | 1.4869 | 19.9036 | 51.398 |
102
+ | 1.4603 | 35.0 | 9870 | 1.4856 | 19.9575 | 51.352 |
103
+ | 1.4558 | 36.0 | 10152 | 1.4845 | 19.9513 | 51.4833 |
104
+ | 1.4558 | 37.0 | 10434 | 1.4840 | 20.0177 | 51.3027 |
105
+ | 1.4486 | 38.0 | 10716 | 1.4833 | 20.0644 | 51.484 |
106
+ | 1.4486 | 39.0 | 10998 | 1.4830 | 20.1001 | 51.5747 |
107
+ | 1.4452 | 40.0 | 11280 | 1.4830 | 20.1194 | 51.456 |
108
+
109
+
110
+ ### Framework versions
111
+
112
+ - Transformers 4.22.1
113
+ - Pytorch 1.12.1
114
+ - Datasets 2.5.1
115
+ - Tokenizers 0.11.0