File size: 4,313 Bytes
132f462 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: apac_5sents_XLS-R_2
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. -->
# apac_5sents_XLS-R_2
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 274.3813
- Wer: 1.0
## 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: 1e-06
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 5000
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:---:|
| 281.5958 | 5.54 | 100 | 273.7941 | 1.0 |
| 281.366 | 11.11 | 200 | 269.9106 | 1.0 |
| 273.1232 | 16.65 | 300 | 263.4907 | 1.0 |
| 261.9693 | 22.22 | 400 | 249.4823 | 1.0 |
| 219.9941 | 27.76 | 500 | 176.5986 | 1.0 |
| 147.673 | 33.32 | 600 | 122.9458 | 1.0 |
| 114.1416 | 38.86 | 700 | 101.6788 | 1.0 |
| 99.6196 | 44.43 | 800 | 91.0054 | 1.0 |
| 90.5221 | 49.97 | 900 | 84.6282 | 1.0 |
| 85.6017 | 55.54 | 1000 | 80.2764 | 1.0 |
| 81.5146 | 61.11 | 1100 | 77.0657 | 1.0 |
| 77.7573 | 66.65 | 1200 | 74.5696 | 1.0 |
| 76.1933 | 72.22 | 1300 | 72.5473 | 1.0 |
| 73.6424 | 77.76 | 1400 | 70.8357 | 1.0 |
| 72.4248 | 83.32 | 1500 | 69.3329 | 1.0 |
| 70.3529 | 88.86 | 1600 | 67.9791 | 1.0 |
| 69.4685 | 94.43 | 1700 | 66.7548 | 1.0 |
| 68.0717 | 99.97 | 1800 | 65.6130 | 1.0 |
| 67.2376 | 105.54 | 1900 | 64.5488 | 1.0 |
| 66.0277 | 111.11 | 2000 | 63.5492 | 1.0 |
| 64.7994 | 116.65 | 2100 | 62.6000 | 1.0 |
| 64.0574 | 122.22 | 2200 | 61.7027 | 1.0 |
| 62.9425 | 127.76 | 2300 | 60.8545 | 1.0 |
| 62.254 | 133.32 | 2400 | 60.0502 | 1.0 |
| 61.3336 | 138.86 | 2500 | 59.2789 | 1.0 |
| 60.7559 | 144.43 | 2600 | 58.5486 | 1.0 |
| 59.7298 | 149.97 | 2700 | 57.8531 | 1.0 |
| 59.4008 | 155.54 | 2800 | 57.1906 | 1.0 |
| 58.6672 | 161.11 | 2900 | 56.5612 | 1.0 |
| 57.6835 | 166.65 | 3000 | 55.9709 | 1.0 |
| 57.4515 | 172.22 | 3100 | 55.4099 | 1.0 |
| 56.7266 | 177.76 | 3200 | 54.8743 | 1.0 |
| 56.3382 | 183.32 | 3300 | 54.3800 | 1.0 |
| 55.4538 | 188.86 | 3400 | 53.9105 | 1.0 |
| 55.4566 | 194.43 | 3500 | 53.4741 | 1.0 |
| 54.6733 | 199.97 | 3600 | 53.0646 | 1.0 |
| 54.5231 | 205.54 | 3700 | 52.6835 | 1.0 |
| 54.1944 | 211.11 | 3800 | 52.3376 | 1.0 |
| 53.5359 | 216.65 | 3900 | 52.0120 | 1.0 |
| 53.4527 | 222.22 | 4000 | 51.7225 | 1.0 |
| 53.0497 | 227.76 | 4100 | 51.4579 | 1.0 |
| 52.9911 | 233.32 | 4200 | 51.2190 | 1.0 |
| 52.3869 | 238.86 | 4300 | 51.0154 | 1.0 |
| 52.4158 | 244.43 | 4400 | 50.8332 | 1.0 |
| 52.1746 | 249.97 | 4500 | 50.6797 | 1.0 |
| 52.2056 | 255.54 | 4600 | 50.5553 | 1.0 |
| 52.1142 | 261.11 | 4700 | 50.4548 | 1.0 |
| 51.7802 | 266.65 | 4800 | 50.3846 | 1.0 |
| 51.9224 | 272.22 | 4900 | 50.3413 | 1.0 |
| 51.7253 | 277.76 | 5000 | 50.3258 | 1.0 |
### Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
|