---
license: cc-by-sa-4.0
tags:
- generated_from_trainer
datasets:
- te_dx_jp
model-index:
- name: t5-base-TEDxJP-8front-1body-8rear
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# t5-base-TEDxJP-8front-1body-8rear

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.
It achieves the following results on the evaluation set:
- Loss: 0.4378
- Wer: 0.1696
- Mer: 0.1640
- Wil: 0.2499
- Wip: 0.7501
- Hits: 55848
- Substitutions: 6316
- Deletions: 2423
- Insertions: 2216
- Cer: 0.1327

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 20
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    | Mer    | Wil    | Wip    | Hits  | Substitutions | Deletions | Insertions | Cer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|:-----:|:-------------:|:---------:|:----------:|:------:|
| 0.5935        | 1.0   | 1457  | 0.4683          | 0.2101 | 0.1975 | 0.2870 | 0.7130 | 55138 | 6765          | 2684      | 4118       | 0.1740 |
| 0.5128        | 2.0   | 2914  | 0.4191          | 0.1781 | 0.1719 | 0.2585 | 0.7415 | 55397 | 6370          | 2820      | 2310       | 0.1415 |
| 0.4948        | 3.0   | 4371  | 0.4085          | 0.1738 | 0.1679 | 0.2540 | 0.7460 | 55644 | 6330          | 2613      | 2284       | 0.1367 |
| 0.4164        | 4.0   | 5828  | 0.4065          | 0.1698 | 0.1643 | 0.2497 | 0.7503 | 55778 | 6263          | 2546      | 2159       | 0.1326 |
| 0.3844        | 5.0   | 7285  | 0.4134          | 0.1705 | 0.1646 | 0.2502 | 0.7498 | 55882 | 6295          | 2410      | 2306       | 0.1344 |
| 0.3327        | 6.0   | 8742  | 0.4175          | 0.1706 | 0.1649 | 0.2506 | 0.7494 | 55774 | 6290          | 2523      | 2203       | 0.1338 |
| 0.2931        | 7.0   | 10199 | 0.4262          | 0.1717 | 0.1655 | 0.2511 | 0.7489 | 55899 | 6307          | 2381      | 2399       | 0.1353 |
| 0.2752        | 8.0   | 11656 | 0.4269          | 0.1694 | 0.1639 | 0.2495 | 0.7505 | 55833 | 6291          | 2463      | 2190       | 0.1324 |
| 0.274         | 9.0   | 13113 | 0.4359          | 0.1701 | 0.1643 | 0.2502 | 0.7498 | 55866 | 6311          | 2410      | 2266       | 0.1333 |
| 0.2426        | 10.0  | 14570 | 0.4378          | 0.1696 | 0.1640 | 0.2499 | 0.7501 | 55848 | 6316          | 2423      | 2216       | 0.1327 |


### Framework versions

- Transformers 4.21.2
- Pytorch 1.12.1+cu116
- Datasets 2.4.0
- Tokenizers 0.12.1