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-1body-2context
|
| 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-1body-2context
|
| 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: 0.4968
|
| 20 |
+
- Wer: 0.1969
|
| 21 |
+
- Mer: 0.1895
|
| 22 |
+
- Wil: 0.2801
|
| 23 |
+
- Wip: 0.7199
|
| 24 |
+
- Hits: 55902
|
| 25 |
+
- Substitutions: 6899
|
| 26 |
+
- Deletions: 3570
|
| 27 |
+
- Insertions: 2599
|
| 28 |
+
- Cer: 0.1727
|
| 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 |
+
| 0.7136 | 1.0 | 746 | 0.5716 | 0.2512 | 0.2345 | 0.3279 | 0.6721 | 54430 | 7249 | 4692 | 4731 | 0.2344 |
|
| 61 |
+
| 0.6267 | 2.0 | 1492 | 0.5152 | 0.2088 | 0.2005 | 0.2917 | 0.7083 | 55245 | 6949 | 4177 | 2732 | 0.2009 |
|
| 62 |
+
| 0.5416 | 3.0 | 2238 | 0.4969 | 0.2025 | 0.1948 | 0.2851 | 0.7149 | 55575 | 6871 | 3925 | 2646 | 0.1802 |
|
| 63 |
+
| 0.5223 | 4.0 | 2984 | 0.4915 | 0.1989 | 0.1917 | 0.2816 | 0.7184 | 55652 | 6826 | 3893 | 2481 | 0.1754 |
|
| 64 |
+
| 0.4985 | 5.0 | 3730 | 0.4929 | 0.1991 | 0.1916 | 0.2814 | 0.7186 | 55759 | 6828 | 3784 | 2603 | 0.1753 |
|
| 65 |
+
| 0.4675 | 6.0 | 4476 | 0.4910 | 0.1969 | 0.1897 | 0.2799 | 0.7201 | 55834 | 6859 | 3678 | 2534 | 0.1756 |
|
| 66 |
+
| 0.445 | 7.0 | 5222 | 0.4940 | 0.1955 | 0.1884 | 0.2782 | 0.7218 | 55881 | 6821 | 3669 | 2485 | 0.1712 |
|
| 67 |
+
| 0.4404 | 8.0 | 5968 | 0.4932 | 0.1979 | 0.1903 | 0.2801 | 0.7199 | 55881 | 6828 | 3662 | 2643 | 0.1742 |
|
| 68 |
+
| 0.4525 | 9.0 | 6714 | 0.4951 | 0.1968 | 0.1893 | 0.2799 | 0.7201 | 55939 | 6897 | 3535 | 2632 | 0.1740 |
|
| 69 |
+
| 0.4077 | 10.0 | 7460 | 0.4968 | 0.1969 | 0.1895 | 0.2801 | 0.7199 | 55902 | 6899 | 3570 | 2599 | 0.1727 |
|
| 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
|