update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,98 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- common_voice
|
| 6 |
+
model-index:
|
| 7 |
+
- name: common7
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# common7
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [common7/checkpoint-4000](https://huggingface.co/common7/checkpoint-4000) on the common_voice dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.3425
|
| 19 |
+
- Wer: 0.3494
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 6e-05
|
| 39 |
+
- train_batch_size: 32
|
| 40 |
+
- eval_batch_size: 16
|
| 41 |
+
- seed: 42
|
| 42 |
+
- gradient_accumulation_steps: 4
|
| 43 |
+
- total_train_batch_size: 128
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- lr_scheduler_warmup_steps: 100
|
| 47 |
+
- num_epochs: 123.0
|
| 48 |
+
- mixed_precision_training: Native AMP
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 53 |
+
|:-------------:|:------:|:-----:|:---------------:|:------:|
|
| 54 |
+
| 2.957 | 3.29 | 500 | 2.9503 | 1.0 |
|
| 55 |
+
| 1.7225 | 6.58 | 1000 | 0.8860 | 0.7703 |
|
| 56 |
+
| 1.4907 | 9.86 | 1500 | 0.6555 | 0.6673 |
|
| 57 |
+
| 1.4177 | 13.16 | 2000 | 0.5784 | 0.6076 |
|
| 58 |
+
| 1.3425 | 16.45 | 2500 | 0.5379 | 0.5718 |
|
| 59 |
+
| 1.33 | 19.73 | 3000 | 0.4962 | 0.5245 |
|
| 60 |
+
| 1.4378 | 23.03 | 3500 | 0.4699 | 0.5098 |
|
| 61 |
+
| 1.1894 | 26.31 | 4000 | 0.4527 | 0.4848 |
|
| 62 |
+
| 1.1844 | 29.6 | 4500 | 0.4309 | 0.4651 |
|
| 63 |
+
| 1.1795 | 32.89 | 5000 | 0.4131 | 0.4524 |
|
| 64 |
+
| 1.1471 | 36.18 | 5500 | 0.4052 | 0.4435 |
|
| 65 |
+
| 1.1337 | 39.47 | 6000 | 0.3927 | 0.4363 |
|
| 66 |
+
| 1.1896 | 42.76 | 6500 | 0.3811 | 0.4254 |
|
| 67 |
+
| 1.1847 | 46.05 | 7000 | 0.3855 | 0.4129 |
|
| 68 |
+
| 0.9954 | 49.34 | 7500 | 0.3729 | 0.3981 |
|
| 69 |
+
| 1.0293 | 52.63 | 8000 | 0.3637 | 0.4014 |
|
| 70 |
+
| 1.0224 | 55.92 | 8500 | 0.3578 | 0.3885 |
|
| 71 |
+
| 1.012 | 59.21 | 9000 | 0.3629 | 0.3930 |
|
| 72 |
+
| 1.0772 | 62.5 | 9500 | 0.3635 | 0.3906 |
|
| 73 |
+
| 1.0344 | 65.79 | 10000 | 0.3469 | 0.3771 |
|
| 74 |
+
| 0.9457 | 69.08 | 10500 | 0.3435 | 0.3735 |
|
| 75 |
+
| 0.9307 | 72.37 | 11000 | 0.3519 | 0.3762 |
|
| 76 |
+
| 0.9523 | 75.65 | 11500 | 0.3443 | 0.3666 |
|
| 77 |
+
| 0.9523 | 78.94 | 12000 | 0.3502 | 0.3757 |
|
| 78 |
+
| 0.9475 | 82.24 | 12500 | 0.3509 | 0.3643 |
|
| 79 |
+
| 0.9971 | 85.52 | 13000 | 0.3502 | 0.3626 |
|
| 80 |
+
| 0.9058 | 88.81 | 13500 | 0.3472 | 0.3605 |
|
| 81 |
+
| 0.8922 | 92.1 | 14000 | 0.3530 | 0.3618 |
|
| 82 |
+
| 0.9 | 95.39 | 14500 | 0.3500 | 0.3574 |
|
| 83 |
+
| 0.9051 | 98.68 | 15000 | 0.3456 | 0.3535 |
|
| 84 |
+
| 0.9304 | 101.97 | 15500 | 0.3438 | 0.3578 |
|
| 85 |
+
| 0.9433 | 105.26 | 16000 | 0.3396 | 0.3530 |
|
| 86 |
+
| 0.8988 | 108.55 | 16500 | 0.3436 | 0.3539 |
|
| 87 |
+
| 0.8789 | 111.84 | 17000 | 0.3426 | 0.3516 |
|
| 88 |
+
| 0.8667 | 115.13 | 17500 | 0.3438 | 0.3506 |
|
| 89 |
+
| 0.8895 | 118.42 | 18000 | 0.3434 | 0.3503 |
|
| 90 |
+
| 0.8888 | 121.71 | 18500 | 0.3425 | 0.3494 |
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
### Framework versions
|
| 94 |
+
|
| 95 |
+
- Transformers 4.17.0.dev0
|
| 96 |
+
- Pytorch 1.10.2
|
| 97 |
+
- Datasets 1.18.3.dev0
|
| 98 |
+
- Tokenizers 0.10.3
|