update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- amazon_reviews_multi
|
7 |
+
metrics:
|
8 |
+
- rouge
|
9 |
+
model-index:
|
10 |
+
- name: mt5-small-finetuned-amazon-en-es
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Sequence-to-sequence Language Modeling
|
14 |
+
type: text2text-generation
|
15 |
+
dataset:
|
16 |
+
name: amazon_reviews_multi
|
17 |
+
type: amazon_reviews_multi
|
18 |
+
config: en
|
19 |
+
split: validation
|
20 |
+
args: en
|
21 |
+
metrics:
|
22 |
+
- name: Rouge1
|
23 |
+
type: rouge
|
24 |
+
value: 19.7803
|
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 |
+
# mt5-small-finetuned-amazon-en-es
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the amazon_reviews_multi dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 3.3776
|
35 |
+
- Rouge1: 19.7803
|
36 |
+
- Rouge2: 11.1015
|
37 |
+
- Rougel: 19.2549
|
38 |
+
- Rougelsum: 19.0799
|
39 |
+
|
40 |
+
## Model description
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Intended uses & limitations
|
45 |
+
|
46 |
+
More information needed
|
47 |
+
|
48 |
+
## Training and evaluation data
|
49 |
+
|
50 |
+
More information needed
|
51 |
+
|
52 |
+
## Training procedure
|
53 |
+
|
54 |
+
### Training hyperparameters
|
55 |
+
|
56 |
+
The following hyperparameters were used during training:
|
57 |
+
- learning_rate: 5.6e-05
|
58 |
+
- train_batch_size: 12
|
59 |
+
- eval_batch_size: 12
|
60 |
+
- seed: 42
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- num_epochs: 16
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
68 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
|
69 |
+
| 2.7004 | 1.0 | 514 | 3.2936 | 17.2056 | 9.3595 | 17.2163 | 16.8991 |
|
70 |
+
| 2.6472 | 2.0 | 1028 | 3.2995 | 17.4081 | 9.9623 | 17.3287 | 16.9701 |
|
71 |
+
| 2.6173 | 3.0 | 1542 | 3.3039 | 17.9581 | 9.5537 | 17.7729 | 17.6022 |
|
72 |
+
| 2.5812 | 4.0 | 2056 | 3.3057 | 17.438 | 8.8331 | 17.1341 | 17.0436 |
|
73 |
+
| 2.5446 | 5.0 | 2570 | 3.3316 | 18.6351 | 10.4851 | 18.5086 | 18.2872 |
|
74 |
+
| 2.5133 | 6.0 | 3084 | 3.3255 | 19.1331 | 10.3396 | 18.8557 | 18.6551 |
|
75 |
+
| 2.4885 | 7.0 | 3598 | 3.3522 | 18.998 | 10.1323 | 18.709 | 18.5662 |
|
76 |
+
| 2.464 | 8.0 | 4112 | 3.3314 | 18.8978 | 10.1726 | 18.5683 | 18.3049 |
|
77 |
+
| 2.4431 | 9.0 | 4626 | 3.3508 | 18.777 | 10.3018 | 18.5103 | 18.2069 |
|
78 |
+
| 2.4092 | 10.0 | 5140 | 3.3520 | 19.3359 | 11.2355 | 19.0265 | 18.7083 |
|
79 |
+
| 2.4082 | 11.0 | 5654 | 3.3534 | 19.3633 | 11.2181 | 19.0465 | 18.7726 |
|
80 |
+
| 2.3815 | 12.0 | 6168 | 3.3687 | 18.7702 | 10.395 | 18.5383 | 18.2271 |
|
81 |
+
| 2.3706 | 13.0 | 6682 | 3.3716 | 19.0868 | 10.6534 | 18.791 | 18.5447 |
|
82 |
+
| 2.3628 | 14.0 | 7196 | 3.3756 | 19.1222 | 10.9791 | 18.6601 | 18.5131 |
|
83 |
+
| 2.3518 | 15.0 | 7710 | 3.3831 | 19.9227 | 11.0903 | 19.4883 | 19.2418 |
|
84 |
+
| 2.3497 | 16.0 | 8224 | 3.3776 | 19.7803 | 11.1015 | 19.2549 | 19.0799 |
|
85 |
+
|
86 |
+
|
87 |
+
### Framework versions
|
88 |
+
|
89 |
+
- Transformers 4.28.1
|
90 |
+
- Pytorch 2.0.0+cu118
|
91 |
+
- Datasets 2.11.0
|
92 |
+
- Tokenizers 0.13.3
|