Hugging Face's logo

aynumosir
/
mt5-base-ainu

Text2Text Generation
Transformers
Safetensors
mt5
Generated from Trainer
Model card Files Community
rigarashi commited on
Commit
7375c26
·
verified ·
1 Parent(s): 5222a8f

Model save

Browse files
Files changed (1) hide show
  1. README.md +29 -29
README.md CHANGED
@@ -1,18 +1,18 @@
1
  ---
 
2
  base_model: google/mt5-small
 
 
3
  datasets:
4
  - arrow
5
- license: apache-2.0
6
  metrics:
7
  - bleu
8
- tags:
9
- - generated_from_trainer
10
  model-index:
11
  - name: mt5-base-ainu
12
  results:
13
  - task:
14
- type: text2text-generation
15
  name: Sequence-to-sequence Language Modeling
 
16
  dataset:
17
  name: arrow
18
  type: arrow
@@ -20,9 +20,9 @@ model-index:
20
  split: None
21
  args: default
22
  metrics:
23
- - type: bleu
24
- value: 34.75882910529557
25
- name: Bleu
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the arrow dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.3717
36
- - Bleu: 34.7588
37
 
38
  ## Model description
39
 
@@ -67,26 +67,26 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Bleu |
69
  |:-------------:|:-------:|:------:|:---------------:|:-------:|
70
- | 2.7342 | 0.9999 | 9341 | 2.2952 | 8.7599 |
71
- | 2.0583 | 2.0 | 18683 | 1.7903 | 19.5422 |
72
- | 1.8006 | 2.9999 | 28024 | 1.6075 | 24.0648 |
73
- | 1.6417 | 4.0 | 37366 | 1.5050 | 27.5308 |
74
- | 1.516 | 4.9999 | 46707 | 1.4466 | 28.5774 |
75
- | 1.4319 | 6.0 | 56049 | 1.4077 | 29.7452 |
76
- | 1.339 | 6.9999 | 65390 | 1.3762 | 30.7138 |
77
- | 1.2797 | 8.0 | 74732 | 1.3575 | 31.1331 |
78
- | 1.2266 | 8.9999 | 84073 | 1.3404 | 31.8717 |
79
- | 1.1595 | 10.0 | 93415 | 1.3375 | 32.4945 |
80
- | 1.1193 | 10.9999 | 102756 | 1.3315 | 32.6273 |
81
- | 1.0606 | 12.0 | 112098 | 1.3252 | 33.4770 |
82
- | 1.0273 | 12.9999 | 121439 | 1.3216 | 33.7973 |
83
- | 0.982 | 14.0 | 130781 | 1.3328 | 33.9583 |
84
- | 0.9462 | 14.9999 | 140122 | 1.3364 | 33.9590 |
85
- | 0.9033 | 16.0 | 149464 | 1.3472 | 34.1416 |
86
- | 0.8785 | 16.9999 | 158805 | 1.3499 | 34.3651 |
87
- | 0.8484 | 18.0 | 168147 | 1.3571 | 34.7063 |
88
- | 0.815 | 18.9999 | 177488 | 1.3641 | 34.6037 |
89
- | 0.7957 | 19.9989 | 186820 | 1.3717 | 34.7588 |
90
 
91
 
92
  ### Framework versions
 
1
  ---
2
+ license: apache-2.0
3
  base_model: google/mt5-small
4
+ tags:
5
+ - generated_from_trainer
6
  datasets:
7
  - arrow
 
8
  metrics:
9
  - bleu
 
 
10
  model-index:
11
  - name: mt5-base-ainu
12
  results:
13
  - task:
 
14
  name: Sequence-to-sequence Language Modeling
15
+ type: text2text-generation
16
  dataset:
17
  name: arrow
18
  type: arrow
 
20
  split: None
21
  args: default
22
  metrics:
23
+ - name: Bleu
24
+ type: bleu
25
+ value: 35.096583697771194
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the arrow dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 1.3701
36
+ - Bleu: 35.0966
37
 
38
  ## Model description
39
 
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Bleu |
69
  |:-------------:|:-------:|:------:|:---------------:|:-------:|
70
+ | 2.73 | 0.9999 | 9341 | 2.2938 | 11.6604 |
71
+ | 2.0536 | 2.0 | 18683 | 1.7883 | 19.8648 |
72
+ | 1.7955 | 2.9999 | 28024 | 1.6060 | 24.9245 |
73
+ | 1.6331 | 4.0 | 37366 | 1.5038 | 27.6152 |
74
+ | 1.5107 | 4.9999 | 46707 | 1.4442 | 28.4249 |
75
+ | 1.4255 | 6.0 | 56049 | 1.4058 | 30.0986 |
76
+ | 1.3373 | 6.9999 | 65390 | 1.3769 | 30.9550 |
77
+ | 1.2758 | 8.0 | 74732 | 1.3546 | 31.6075 |
78
+ | 1.2259 | 8.9999 | 84073 | 1.3380 | 32.0155 |
79
+ | 1.1555 | 10.0 | 93415 | 1.3318 | 32.5095 |
80
+ | 1.1166 | 10.9999 | 102756 | 1.3263 | 32.9619 |
81
+ | 1.0564 | 12.0 | 112098 | 1.3220 | 33.5983 |
82
+ | 1.0222 | 12.9999 | 121439 | 1.3162 | 33.7293 |
83
+ | 0.9764 | 14.0 | 130781 | 1.3304 | 34.0143 |
84
+ | 0.9472 | 14.9999 | 140122 | 1.3353 | 34.3792 |
85
+ | 0.9034 | 16.0 | 149464 | 1.3429 | 34.4296 |
86
+ | 0.8732 | 16.9999 | 158805 | 1.3470 | 34.5817 |
87
+ | 0.8465 | 18.0 | 168147 | 1.3554 | 34.9795 |
88
+ | 0.8171 | 18.9999 | 177488 | 1.3618 | 34.9659 |
89
+ | 0.7912 | 19.9989 | 186820 | 1.3701 | 35.0966 |
90
 
91
 
92
  ### Framework versions