Commit
•
dc7ef6f
1
Parent(s):
7556298
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-large-xls-r-300m-hi-test123
|
| 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 |
+
# wav2vec2-large-xls-r-300m-hi-test123
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.6588
|
| 20 |
+
- Wer: 0.2987
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.000245
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 2000
|
| 48 |
+
- num_epochs: 60
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 12.809 | 1.36 | 200 | 6.2066 | 1.0 |
|
| 56 |
+
| 4.3402 | 2.72 | 400 | 3.5184 | 1.0 |
|
| 57 |
+
| 3.4365 | 4.08 | 600 | 3.2779 | 1.0 |
|
| 58 |
+
| 1.8643 | 5.44 | 800 | 0.9875 | 0.6270 |
|
| 59 |
+
| 0.7504 | 6.8 | 1000 | 0.6382 | 0.4666 |
|
| 60 |
+
| 0.5328 | 8.16 | 1200 | 0.6075 | 0.4505 |
|
| 61 |
+
| 0.4364 | 9.52 | 1400 | 0.5785 | 0.4215 |
|
| 62 |
+
| 0.3777 | 10.88 | 1600 | 0.6279 | 0.4227 |
|
| 63 |
+
| 0.3374 | 12.24 | 1800 | 0.6536 | 0.4192 |
|
| 64 |
+
| 0.3236 | 13.6 | 2000 | 0.5911 | 0.4047 |
|
| 65 |
+
| 0.2877 | 14.96 | 2200 | 0.5955 | 0.4097 |
|
| 66 |
+
| 0.2643 | 16.33 | 2400 | 0.5923 | 0.3744 |
|
| 67 |
+
| 0.2421 | 17.68 | 2600 | 0.6307 | 0.3814 |
|
| 68 |
+
| 0.2218 | 19.05 | 2800 | 0.6036 | 0.3764 |
|
| 69 |
+
| 0.2046 | 20.41 | 3000 | 0.6286 | 0.3797 |
|
| 70 |
+
| 0.191 | 21.77 | 3200 | 0.6517 | 0.3889 |
|
| 71 |
+
| 0.1856 | 23.13 | 3400 | 0.6193 | 0.3661 |
|
| 72 |
+
| 0.1721 | 24.49 | 3600 | 0.7034 | 0.3727 |
|
| 73 |
+
| 0.1656 | 25.85 | 3800 | 0.6293 | 0.3591 |
|
| 74 |
+
| 0.1532 | 27.21 | 4000 | 0.6075 | 0.3611 |
|
| 75 |
+
| 0.1507 | 28.57 | 4200 | 0.6313 | 0.3565 |
|
| 76 |
+
| 0.1381 | 29.93 | 4400 | 0.6564 | 0.3578 |
|
| 77 |
+
| 0.1359 | 31.29 | 4600 | 0.6724 | 0.3543 |
|
| 78 |
+
| 0.1248 | 32.65 | 4800 | 0.6789 | 0.3512 |
|
| 79 |
+
| 0.1198 | 34.01 | 5000 | 0.6442 | 0.3539 |
|
| 80 |
+
| 0.1125 | 35.37 | 5200 | 0.6676 | 0.3419 |
|
| 81 |
+
| 0.1036 | 36.73 | 5400 | 0.7017 | 0.3435 |
|
| 82 |
+
| 0.0982 | 38.09 | 5600 | 0.6828 | 0.3319 |
|
| 83 |
+
| 0.0971 | 39.45 | 5800 | 0.6112 | 0.3351 |
|
| 84 |
+
| 0.0968 | 40.81 | 6000 | 0.6424 | 0.3252 |
|
| 85 |
+
| 0.0893 | 42.18 | 6200 | 0.6707 | 0.3304 |
|
| 86 |
+
| 0.0878 | 43.54 | 6400 | 0.6432 | 0.3236 |
|
| 87 |
+
| 0.0827 | 44.89 | 6600 | 0.6696 | 0.3240 |
|
| 88 |
+
| 0.0788 | 46.26 | 6800 | 0.6564 | 0.3180 |
|
| 89 |
+
| 0.0753 | 47.62 | 7000 | 0.6574 | 0.3130 |
|
| 90 |
+
| 0.0674 | 48.98 | 7200 | 0.6698 | 0.3175 |
|
| 91 |
+
| 0.0676 | 50.34 | 7400 | 0.6441 | 0.3142 |
|
| 92 |
+
| 0.0626 | 51.7 | 7600 | 0.6642 | 0.3121 |
|
| 93 |
+
| 0.0617 | 53.06 | 7800 | 0.6615 | 0.3117 |
|
| 94 |
+
| 0.0599 | 54.42 | 8000 | 0.6634 | 0.3059 |
|
| 95 |
+
| 0.0538 | 55.78 | 8200 | 0.6464 | 0.3033 |
|
| 96 |
+
| 0.0571 | 57.14 | 8400 | 0.6503 | 0.3018 |
|
| 97 |
+
| 0.0491 | 58.5 | 8600 | 0.6625 | 0.3025 |
|
| 98 |
+
| 0.0511 | 59.86 | 8800 | 0.6588 | 0.2987 |
|
| 99 |
+
|
| 100 |
+
|
| 101 |
+
### Framework versions
|
| 102 |
+
|
| 103 |
+
- Transformers 4.16.2
|
| 104 |
+
- Pytorch 1.10.0+cu111
|
| 105 |
+
- Datasets 1.18.3
|
| 106 |
+
- Tokenizers 0.11.0
|