File size: 3,625 Bytes
180d66b |
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 |
---
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2_XLSR_darija_maroc
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. -->
# wav2vec2_XLSR_darija_maroc
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: 0.2860
- Wer: 0.3290
## 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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 4.9354 | 0.83 | 400 | 2.0492 | 1.0371 |
| 0.8236 | 1.66 | 800 | 0.4434 | 0.5832 |
| 0.4821 | 2.49 | 1200 | 0.3597 | 0.5114 |
| 0.3823 | 3.32 | 1600 | 0.3265 | 0.4758 |
| 0.3231 | 4.15 | 2000 | 0.3149 | 0.4526 |
| 0.2854 | 4.97 | 2400 | 0.2797 | 0.4237 |
| 0.2529 | 5.8 | 2800 | 0.3027 | 0.4415 |
| 0.2493 | 6.63 | 3200 | 0.2926 | 0.4264 |
| 0.2138 | 7.46 | 3600 | 0.2857 | 0.4169 |
| 0.2067 | 8.29 | 4000 | 0.2743 | 0.4099 |
| 0.1898 | 9.12 | 4400 | 0.2798 | 0.3993 |
| 0.1755 | 9.95 | 4800 | 0.2800 | 0.3913 |
| 0.1603 | 10.78 | 5200 | 0.2709 | 0.3860 |
| 0.1608 | 11.61 | 5600 | 0.2716 | 0.3872 |
| 0.1462 | 12.44 | 6000 | 0.2697 | 0.3825 |
| 0.137 | 13.26 | 6400 | 0.2855 | 0.3819 |
| 0.1326 | 14.09 | 6800 | 0.2860 | 0.3733 |
| 0.123 | 14.92 | 7200 | 0.2677 | 0.3813 |
| 0.1168 | 15.75 | 7600 | 0.2780 | 0.3740 |
| 0.1113 | 16.58 | 8000 | 0.2926 | 0.3719 |
| 0.1057 | 17.41 | 8400 | 0.2927 | 0.3704 |
| 0.0996 | 18.24 | 8800 | 0.2825 | 0.3602 |
| 0.0967 | 19.07 | 9200 | 0.2983 | 0.3641 |
| 0.0925 | 19.9 | 9600 | 0.2843 | 0.3576 |
| 0.0894 | 20.73 | 10000 | 0.2726 | 0.3668 |
| 0.0836 | 21.55 | 10400 | 0.2829 | 0.3560 |
| 0.0789 | 22.38 | 10800 | 0.2806 | 0.3508 |
| 0.0778 | 23.21 | 11200 | 0.2849 | 0.3540 |
| 0.0742 | 24.04 | 11600 | 0.2770 | 0.3436 |
| 0.0679 | 24.87 | 12000 | 0.2850 | 0.3425 |
| 0.063 | 25.7 | 12400 | 0.2846 | 0.3366 |
| 0.0593 | 26.53 | 12800 | 0.2811 | 0.3351 |
| 0.0586 | 27.36 | 13200 | 0.2863 | 0.3322 |
| 0.0555 | 28.19 | 13600 | 0.2819 | 0.3311 |
| 0.053 | 29.02 | 14000 | 0.2874 | 0.3301 |
| 0.0498 | 29.84 | 14400 | 0.2860 | 0.3290 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
|