update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-sa-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- te_dx_jp
|
| 7 |
+
model-index:
|
| 8 |
+
- name: t5-base-TEDxJP-11body-0context
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# t5-base-TEDxJP-11body-0context
|
| 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: 0.8068
|
| 20 |
+
- Wer: 0.1976
|
| 21 |
+
- Mer: 0.1904
|
| 22 |
+
- Wil: 0.2816
|
| 23 |
+
- Wip: 0.7184
|
| 24 |
+
- Hits: 602335
|
| 25 |
+
- Substitutions: 75050
|
| 26 |
+
- Deletions: 39435
|
| 27 |
+
- Insertions: 27185
|
| 28 |
+
- Cer: 0.1625
|
| 29 |
+
|
| 30 |
+
## Model description
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Intended uses & limitations
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training and evaluation data
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Training procedure
|
| 43 |
+
|
| 44 |
+
### Training hyperparameters
|
| 45 |
+
|
| 46 |
+
The following hyperparameters were used during training:
|
| 47 |
+
- learning_rate: 0.0001
|
| 48 |
+
- train_batch_size: 64
|
| 49 |
+
- eval_batch_size: 8
|
| 50 |
+
- seed: 42
|
| 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
|
| 54 |
+
- num_epochs: 10
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Mer | Wil | Wip | Hits | Substitutions | Deletions | Insertions | Cer |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:------:|:------:|:-------------:|:---------:|:----------:|:------:|
|
| 60 |
+
| 0.8909 | 1.0 | 746 | 0.7722 | 0.3120 | 0.2861 | 0.3989 | 0.6011 | 558138 | 99887 | 58795 | 64983 | 0.2652 |
|
| 61 |
+
| 0.6786 | 2.0 | 1492 | 0.7021 | 0.2226 | 0.2122 | 0.3069 | 0.6931 | 592242 | 78773 | 45805 | 34978 | 0.1862 |
|
| 62 |
+
| 0.5627 | 3.0 | 2238 | 0.6996 | 0.2104 | 0.2016 | 0.2942 | 0.7058 | 597381 | 76593 | 42846 | 31392 | 0.1752 |
|
| 63 |
+
| 0.489 | 4.0 | 2984 | 0.7161 | 0.2030 | 0.1952 | 0.2865 | 0.7135 | 599808 | 75155 | 41857 | 28506 | 0.1684 |
|
| 64 |
+
| 0.4355 | 5.0 | 3730 | 0.7389 | 0.2000 | 0.1924 | 0.2837 | 0.7163 | 601815 | 75247 | 39758 | 28335 | 0.1651 |
|
| 65 |
+
| 0.3836 | 6.0 | 4476 | 0.7537 | 0.1992 | 0.1918 | 0.2829 | 0.7171 | 601846 | 75046 | 39928 | 27815 | 0.1640 |
|
| 66 |
+
| 0.3617 | 7.0 | 5222 | 0.7743 | 0.1995 | 0.1918 | 0.2832 | 0.7168 | 602287 | 75268 | 39265 | 28445 | 0.1642 |
|
| 67 |
+
| 0.3258 | 8.0 | 5968 | 0.7907 | 0.1971 | 0.1899 | 0.2809 | 0.7191 | 602800 | 74887 | 39133 | 27258 | 0.1620 |
|
| 68 |
+
| 0.3225 | 9.0 | 6714 | 0.8035 | 0.1981 | 0.1908 | 0.2823 | 0.7177 | 602418 | 75372 | 39030 | 27625 | 0.1630 |
|
| 69 |
+
| 0.3162 | 10.0 | 7460 | 0.8068 | 0.1976 | 0.1904 | 0.2816 | 0.7184 | 602335 | 75050 | 39435 | 27185 | 0.1625 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- Transformers 4.12.5
|
| 75 |
+
- Pytorch 1.10.0+cu102
|
| 76 |
+
- Datasets 1.15.1
|
| 77 |
+
- Tokenizers 0.10.3
|