Hugging Face's logo

Padomin
/
t5-base-TEDxJP-9front-1body-9rear

Text2Text Generation
Transformers
PyTorch
TensorBoard
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
Padomin commited on
Commit
abaa9d0
·
1 Parent(s): 3e5ba47

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -21
README.md CHANGED
@@ -16,16 +16,16 @@ should probably proofread and complete it, then remove this comment. -->
16
 
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: 0.4380
20
- - Wer: 0.1702
21
- - Mer: 0.1642
22
- - Wil: 0.2491
23
- - Wip: 0.7509
24
- - Hits: 55961
25
- - Substitutions: 6246
26
- - Deletions: 2380
27
- - Insertions: 2365
28
- - Cer: 0.1359
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
- | 0.5921 | 1.0 | 1457 | 0.4603 | 0.2169 | 0.2032 | 0.2904 | 0.7096 | 54918 | 6555 | 3114 | 4338 | 0.1908 |
61
- | 0.4981 | 2.0 | 2914 | 0.4174 | 0.1801 | 0.1730 | 0.2603 | 0.7397 | 55608 | 6467 | 2512 | 2652 | 0.1428 |
62
- | 0.4609 | 3.0 | 4371 | 0.4105 | 0.1718 | 0.1659 | 0.2518 | 0.7482 | 55794 | 6324 | 2469 | 2304 | 0.1342 |
63
- | 0.3985 | 4.0 | 5828 | 0.4066 | 0.1699 | 0.1644 | 0.2494 | 0.7506 | 55781 | 6237 | 2569 | 2169 | 0.1346 |
64
- | 0.3875 | 5.0 | 7285 | 0.4110 | 0.1709 | 0.1653 | 0.2505 | 0.7495 | 55753 | 6257 | 2577 | 2206 | 0.1353 |
65
- | 0.3261 | 6.0 | 8742 | 0.4166 | 0.1691 | 0.1638 | 0.2491 | 0.7509 | 55752 | 6256 | 2579 | 2085 | 0.1350 |
66
- | 0.2923 | 7.0 | 10199 | 0.4238 | 0.1695 | 0.1639 | 0.2495 | 0.7505 | 55859 | 6292 | 2436 | 2222 | 0.1354 |
67
- | 0.2629 | 8.0 | 11656 | 0.4292 | 0.1706 | 0.1645 | 0.2498 | 0.7502 | 55948 | 6270 | 2369 | 2380 | 0.1368 |
68
- | 0.2862 | 9.0 | 13113 | 0.4337 | 0.1697 | 0.1636 | 0.2481 | 0.7519 | 56005 | 6203 | 2379 | 2376 | 0.1355 |
69
- | 0.2562 | 10.0 | 14570 | 0.4380 | 0.1702 | 0.1642 | 0.2491 | 0.7509 | 55961 | 6246 | 2380 | 2365 | 0.1359 |
70
 
71
 
72
  ### Framework versions
 
16
 
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: 0.4361
20
+ - Wer: 0.1687
21
+ - Mer: 0.1630
22
+ - Wil: 0.2486
23
+ - Wip: 0.7514
24
+ - Hits: 55941
25
+ - Substitutions: 6292
26
+ - Deletions: 2354
27
+ - Insertions: 2252
28
+ - Cer: 0.1338
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
+ | 0.6124 | 1.0 | 1457 | 0.4613 | 0.2407 | 0.2209 | 0.3091 | 0.6909 | 54843 | 6758 | 2986 | 5804 | 0.2153 |
61
+ | 0.4968 | 2.0 | 2914 | 0.4171 | 0.1777 | 0.1716 | 0.2580 | 0.7420 | 55404 | 6354 | 2829 | 2293 | 0.1402 |
62
+ | 0.4817 | 3.0 | 4371 | 0.4129 | 0.1731 | 0.1673 | 0.2534 | 0.7466 | 55636 | 6332 | 2619 | 2227 | 0.1349 |
63
+ | 0.4257 | 4.0 | 5828 | 0.4089 | 0.1722 | 0.1659 | 0.2520 | 0.7480 | 55904 | 6346 | 2337 | 2437 | 0.1361 |
64
+ | 0.3831 | 5.0 | 7285 | 0.4144 | 0.1705 | 0.1646 | 0.2508 | 0.7492 | 55868 | 6343 | 2376 | 2290 | 0.1358 |
65
+ | 0.3057 | 6.0 | 8742 | 0.4198 | 0.1690 | 0.1632 | 0.2492 | 0.7508 | 55972 | 6333 | 2282 | 2298 | 0.1350 |
66
+ | 0.2919 | 7.0 | 10199 | 0.4220 | 0.1693 | 0.1635 | 0.2492 | 0.7508 | 55936 | 6310 | 2341 | 2281 | 0.1337 |
67
+ | 0.2712 | 8.0 | 11656 | 0.4252 | 0.1688 | 0.1632 | 0.2487 | 0.7513 | 55905 | 6286 | 2396 | 2218 | 0.1348 |
68
+ | 0.2504 | 9.0 | 13113 | 0.4332 | 0.1685 | 0.1629 | 0.2482 | 0.7518 | 55931 | 6270 | 2386 | 2226 | 0.1331 |
69
+ | 0.2446 | 10.0 | 14570 | 0.4361 | 0.1687 | 0.1630 | 0.2486 | 0.7514 | 55941 | 6292 | 2354 | 2252 | 0.1338 |
70
 
71
 
72
  ### Framework versions