Hugging Face's logo

Padomin
/
t5-base-TEDxJP-10front-1body-10rear

Text2Text Generation
Transformers
PyTorch
TensorBoard
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
Padomin commited on
Commit
4b3af82
·
1 Parent(s): 267bfe9

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -20
README.md CHANGED
@@ -17,15 +17,15 @@ should probably proofread and complete it, then remove this comment. -->
17
  This model is a fine-tuned version of [sonoisa/t5-base-japanese](https://huggingface.co/sonoisa/t5-base-japanese) on the te_dx_jp dataset.
18
  It achieves the following results on the evaluation set:
19
  - Loss: 0.4366
20
- - Wer: 0.1686
21
- - Mer: 0.1630
22
- - Wil: 0.2490
23
- - Wip: 0.7510
24
- - Hits: 55913
25
- - Substitutions: 6325
26
- - Deletions: 2349
27
- - Insertions: 2213
28
- - Cer: 0.1324
29
 
30
  ## Model description
31
 
@@ -47,7 +47,7 @@ The following hyperparameters were used during training:
47
  - learning_rate: 0.0001
48
  - train_batch_size: 32
49
  - eval_batch_size: 32
50
- - seed: 30
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_ratio: 0.1
@@ -57,16 +57,16 @@ The following hyperparameters were used during training:
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Mer | Wil | Wip | Hits | Substitutions | Deletions | Insertions | Cer |
59
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|:-----:|:-------------:|:---------:|:----------:|:------:|
60
- | 0.5904 | 1.0 | 1457 | 0.4553 | 0.2049 | 0.1941 | 0.2824 | 0.7176 | 54935 | 6595 | 3057 | 3580 | 0.1816 |
61
- | 0.5001 | 2.0 | 2914 | 0.4201 | 0.1858 | 0.1776 | 0.2657 | 0.7343 | 55561 | 6554 | 2472 | 2973 | 0.1501 |
62
- | 0.4615 | 3.0 | 4371 | 0.4099 | 0.1748 | 0.1685 | 0.2544 | 0.7456 | 55706 | 6326 | 2555 | 2410 | 0.1414 |
63
- | 0.3988 | 4.0 | 5828 | 0.4040 | 0.1710 | 0.1654 | 0.2514 | 0.7486 | 55734 | 6319 | 2534 | 2189 | 0.1346 |
64
- | 0.3859 | 5.0 | 7285 | 0.4131 | 0.1689 | 0.1635 | 0.2487 | 0.7513 | 55808 | 6245 | 2534 | 2129 | 0.1327 |
65
- | 0.3259 | 6.0 | 8742 | 0.4138 | 0.1695 | 0.1639 | 0.2508 | 0.7492 | 55837 | 6400 | 2350 | 2198 | 0.1325 |
66
- | 0.2915 | 7.0 | 10199 | 0.4233 | 0.1696 | 0.1637 | 0.2499 | 0.7501 | 55932 | 6344 | 2311 | 2297 | 0.1329 |
67
- | 0.2638 | 8.0 | 11656 | 0.4298 | 0.1689 | 0.1633 | 0.2492 | 0.7508 | 55892 | 6319 | 2376 | 2213 | 0.1325 |
68
- | 0.2888 | 9.0 | 13113 | 0.4321 | 0.1686 | 0.1630 | 0.2492 | 0.7508 | 55909 | 6343 | 2335 | 2210 | 0.1319 |
69
- | 0.2614 | 10.0 | 14570 | 0.4366 | 0.1686 | 0.1630 | 0.2490 | 0.7510 | 55913 | 6325 | 2349 | 2213 | 0.1324 |
70
 
71
 
72
  ### Framework versions
 
17
  This model is a fine-tuned version of [sonoisa/t5-base-japanese](https://huggingface.co/sonoisa/t5-base-japanese) on the te_dx_jp dataset.
18
  It achieves the following results on the evaluation set:
19
  - Loss: 0.4366
20
+ - Wer: 0.1693
21
+ - Mer: 0.1636
22
+ - Wil: 0.2493
23
+ - Wip: 0.7507
24
+ - Hits: 55904
25
+ - Substitutions: 6304
26
+ - Deletions: 2379
27
+ - Insertions: 2249
28
+ - Cer: 0.1332
29
 
30
  ## Model description
31
 
 
47
  - learning_rate: 0.0001
48
  - train_batch_size: 32
49
  - eval_batch_size: 32
50
+ - seed: 40
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_ratio: 0.1
 
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Mer | Wil | Wip | Hits | Substitutions | Deletions | Insertions | Cer |
59
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|:-----:|:-------------:|:---------:|:----------:|:------:|
60
+ | 0.6166 | 1.0 | 1457 | 0.4595 | 0.2096 | 0.1979 | 0.2878 | 0.7122 | 54866 | 6757 | 2964 | 3819 | 0.1793 |
61
+ | 0.4985 | 2.0 | 2914 | 0.4190 | 0.1769 | 0.1710 | 0.2587 | 0.7413 | 55401 | 6467 | 2719 | 2241 | 0.1417 |
62
+ | 0.4787 | 3.0 | 4371 | 0.4130 | 0.1728 | 0.1670 | 0.2534 | 0.7466 | 55677 | 6357 | 2553 | 2249 | 0.1368 |
63
+ | 0.4299 | 4.0 | 5828 | 0.4085 | 0.1726 | 0.1665 | 0.2530 | 0.7470 | 55799 | 6381 | 2407 | 2357 | 0.1348 |
64
+ | 0.3855 | 5.0 | 7285 | 0.4130 | 0.1702 | 0.1644 | 0.2501 | 0.7499 | 55887 | 6309 | 2391 | 2292 | 0.1336 |
65
+ | 0.3109 | 6.0 | 8742 | 0.4182 | 0.1732 | 0.1668 | 0.2525 | 0.7475 | 55893 | 6317 | 2377 | 2494 | 0.1450 |
66
+ | 0.3027 | 7.0 | 10199 | 0.4256 | 0.1691 | 0.1633 | 0.2486 | 0.7514 | 55949 | 6273 | 2365 | 2283 | 0.1325 |
67
+ | 0.2729 | 8.0 | 11656 | 0.4252 | 0.1709 | 0.1649 | 0.2503 | 0.7497 | 55909 | 6283 | 2395 | 2362 | 0.1375 |
68
+ | 0.2531 | 9.0 | 13113 | 0.4329 | 0.1696 | 0.1639 | 0.2499 | 0.7501 | 55870 | 6322 | 2395 | 2235 | 0.1334 |
69
+ | 0.2388 | 10.0 | 14570 | 0.4366 | 0.1693 | 0.1636 | 0.2493 | 0.7507 | 55904 | 6304 | 2379 | 2249 | 0.1332 |
70
 
71
 
72
  ### Framework versions