Hugging Face's logo

jdmartinev
/
MLEAFIT_es2ptT5

Text2Text Generation
Transformers
PyTorch
mt5
Generated from Trainer
Eval Results
Model card Files Community
2
jdmartinev commited on
Commit
5001e9f
·
1 Parent(s): c577fa0

End of training

Browse files
Files changed (2) hide show
  1. README.md +16 -15
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -4,7 +4,7 @@ base_model: t5-small
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
- - opus_books
8
  metrics:
9
  - bleu
10
  model-index:
@@ -14,15 +14,15 @@ model-index:
14
  name: Sequence-to-sequence Language Modeling
15
  type: text2text-generation
16
  dataset:
17
- name: opus_books
18
- type: opus_books
19
  config: es-pt
20
  split: train
21
  args: es-pt
22
  metrics:
23
  - name: Bleu
24
  type: bleu
25
- value: 0.0779
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,11 +30,11 @@ should probably proofread and complete it, then remove this comment. -->
30
 
31
  # MLEAFIT_es2ptT5
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: 3.9572
36
- - Bleu: 0.0779
37
- - Gen Len: 18.1353
38
 
39
  ## Model description
40
 
@@ -54,19 +54,20 @@ More information needed
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 2e-05
57
- - train_batch_size: 32
58
- - eval_batch_size: 32
59
  - seed: 42
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
- - num_epochs: 2
63
 
64
  ### Training results
65
 
66
- | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
67
- |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
68
- | No log | 1.0 | 34 | 4.0394 | 0.0789 | 18.203 |
69
- | No log | 2.0 | 68 | 3.9572 | 0.0779 | 18.1353 |
 
70
 
71
 
72
  ### Framework versions
 
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
+ - tatoeba
8
  metrics:
9
  - bleu
10
  model-index:
 
14
  name: Sequence-to-sequence Language Modeling
15
  type: text2text-generation
16
  dataset:
17
+ name: tatoeba
18
+ type: tatoeba
19
  config: es-pt
20
  split: train
21
  args: es-pt
22
  metrics:
23
  - name: Bleu
24
  type: bleu
25
+ value: 11.2994
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
30
 
31
  # MLEAFIT_es2ptT5
32
 
33
+ This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the tatoeba dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 1.7472
36
+ - Bleu: 11.2994
37
+ - Gen Len: 15.8838
38
 
39
  ## Model description
40
 
 
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 2e-05
57
+ - train_batch_size: 64
58
+ - eval_batch_size: 64
59
  - seed: 42
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
+ - num_epochs: 3
63
 
64
  ### Training results
65
 
66
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
67
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
68
+ | 2.6856 | 1.0 | 858 | 1.9674 | 8.9672 | 15.7279 |
69
+ | 2.1422 | 2.0 | 1716 | 1.7900 | 10.7687 | 15.8897 |
70
+ | 2.0298 | 3.0 | 2574 | 1.7472 | 11.2994 | 15.8838 |
71
 
72
 
73
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8fee1bb76b320c6f968c24c6f3e7f81e594d2a40d9c62e01240d03d0a5ad409f
3
  size 242071641
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cb5b20a6b5108dd431d45bae2d29a802cd4ad76111ba0a360e1f0b7fb449186
3
  size 242071641