marciovbarbosa
commited on
Commit
•
098cb93
1
Parent(s):
68a058b
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- wmt16
|
7 |
+
metrics:
|
8 |
+
- bleu
|
9 |
+
model-index:
|
10 |
+
- name: t5-small-finetuned-de-to-en-lr1e-4
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Sequence-to-sequence Language Modeling
|
14 |
+
type: text2text-generation
|
15 |
+
dataset:
|
16 |
+
name: wmt16
|
17 |
+
type: wmt16
|
18 |
+
args: de-en
|
19 |
+
metrics:
|
20 |
+
- name: Bleu
|
21 |
+
type: bleu
|
22 |
+
value: 11.3891
|
23 |
+
---
|
24 |
+
|
25 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
26 |
+
should probably proofread and complete it, then remove this comment. -->
|
27 |
+
|
28 |
+
# t5-small-finetuned-de-to-en-lr1e-4
|
29 |
+
|
30 |
+
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the wmt16 dataset.
|
31 |
+
It achieves the following results on the evaluation set:
|
32 |
+
- Loss: 1.8982
|
33 |
+
- Bleu: 11.3891
|
34 |
+
- Gen Len: 17.2881
|
35 |
+
|
36 |
+
## Model description
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Intended uses & limitations
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Training and evaluation data
|
45 |
+
|
46 |
+
More information needed
|
47 |
+
|
48 |
+
## Training procedure
|
49 |
+
|
50 |
+
### Training hyperparameters
|
51 |
+
|
52 |
+
The following hyperparameters were used during training:
|
53 |
+
- learning_rate: 0.0001
|
54 |
+
- train_batch_size: 8
|
55 |
+
- eval_batch_size: 8
|
56 |
+
- seed: 42
|
57 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
58 |
+
- lr_scheduler_type: linear
|
59 |
+
- num_epochs: 10
|
60 |
+
|
61 |
+
### Training results
|
62 |
+
|
63 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
64 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
|
65 |
+
| No log | 1.0 | 272 | 2.0257 | 9.1145 | 17.3171 |
|
66 |
+
| 2.4581 | 2.0 | 544 | 1.9562 | 10.3226 | 17.3218 |
|
67 |
+
| 2.4581 | 3.0 | 816 | 1.9339 | 10.5298 | 17.3414 |
|
68 |
+
| 2.1446 | 4.0 | 1088 | 1.9203 | 10.8193 | 17.3398 |
|
69 |
+
| 2.1446 | 5.0 | 1360 | 1.9070 | 10.9929 | 17.2848 |
|
70 |
+
| 1.9797 | 6.0 | 1632 | 1.9036 | 11.2258 | 17.3258 |
|
71 |
+
| 1.9797 | 7.0 | 1904 | 1.8966 | 11.3839 | 17.3044 |
|
72 |
+
| 1.8939 | 8.0 | 2176 | 1.8999 | 11.4197 | 17.2768 |
|
73 |
+
| 1.8939 | 9.0 | 2448 | 1.8971 | 11.4326 | 17.2791 |
|
74 |
+
| 1.8365 | 10.0 | 2720 | 1.8982 | 11.3891 | 17.2881 |
|
75 |
+
|
76 |
+
|
77 |
+
### Framework versions
|
78 |
+
|
79 |
+
- Transformers 4.12.5
|
80 |
+
- Pytorch 1.10.0+cu111
|
81 |
+
- Datasets 1.16.1
|
82 |
+
- Tokenizers 0.10.3
|