wav2vec2-large-xlsr-moroccan-darija-2-finetuned-nejma-6
This model is a fine-tuned version of boumehdi/wav2vec2-large-xlsr-moroccan-darija on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3702
- Accuracy: 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: 3e-06
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.6131 | 1.0 | 4 | 0.6147 | 1.0 |
0.6062 | 2.0 | 8 | 0.6087 | 1.0 |
0.5928 | 3.0 | 12 | 0.6026 | 1.0 |
0.5859 | 4.0 | 16 | 0.5957 | 1.0 |
0.5802 | 5.0 | 20 | 0.5887 | 1.0 |
0.5709 | 6.0 | 24 | 0.5817 | 1.0 |
0.5693 | 7.0 | 28 | 0.5744 | 1.0 |
0.5576 | 8.0 | 32 | 0.5671 | 1.0 |
0.5553 | 9.0 | 36 | 0.5599 | 1.0 |
0.5458 | 10.0 | 40 | 0.5525 | 1.0 |
0.5322 | 11.0 | 44 | 0.5453 | 1.0 |
0.533 | 12.0 | 48 | 0.5379 | 1.0 |
0.5125 | 13.0 | 52 | 0.5306 | 1.0 |
0.5086 | 14.0 | 56 | 0.5236 | 1.0 |
0.503 | 15.0 | 60 | 0.5164 | 1.0 |
0.5014 | 16.0 | 64 | 0.5093 | 1.0 |
0.4885 | 17.0 | 68 | 0.5025 | 1.0 |
0.4823 | 18.0 | 72 | 0.4956 | 1.0 |
0.474 | 19.0 | 76 | 0.4887 | 1.0 |
0.469 | 20.0 | 80 | 0.4822 | 1.0 |
0.4664 | 21.0 | 84 | 0.4757 | 1.0 |
0.4499 | 22.0 | 88 | 0.4694 | 1.0 |
0.4449 | 23.0 | 92 | 0.4631 | 1.0 |
0.4324 | 24.0 | 96 | 0.4569 | 1.0 |
0.4212 | 25.0 | 100 | 0.4509 | 1.0 |
0.4257 | 26.0 | 104 | 0.4451 | 1.0 |
0.4143 | 27.0 | 108 | 0.4395 | 1.0 |
0.4152 | 28.0 | 112 | 0.4357 | 1.0 |
0.4154 | 29.0 | 116 | 0.4308 | 1.0 |
0.3923 | 30.0 | 120 | 0.4258 | 1.0 |
0.4004 | 31.0 | 124 | 0.4208 | 1.0 |
0.39 | 32.0 | 128 | 0.4158 | 1.0 |
0.3896 | 33.0 | 132 | 0.4112 | 1.0 |
0.384 | 34.0 | 136 | 0.4068 | 1.0 |
0.3794 | 35.0 | 140 | 0.4026 | 1.0 |
0.3667 | 36.0 | 144 | 0.3987 | 1.0 |
0.3735 | 37.0 | 148 | 0.3949 | 1.0 |
0.366 | 38.0 | 152 | 0.3914 | 1.0 |
0.3662 | 39.0 | 156 | 0.3882 | 1.0 |
0.3615 | 40.0 | 160 | 0.3851 | 1.0 |
0.363 | 41.0 | 164 | 0.3824 | 1.0 |
0.354 | 42.0 | 168 | 0.3800 | 1.0 |
0.3467 | 43.0 | 172 | 0.3779 | 1.0 |
0.3549 | 44.0 | 176 | 0.3759 | 1.0 |
0.3432 | 45.0 | 180 | 0.3743 | 1.0 |
0.3417 | 46.0 | 184 | 0.3729 | 1.0 |
0.3465 | 47.0 | 188 | 0.3719 | 1.0 |
0.3548 | 48.0 | 192 | 0.3710 | 1.0 |
0.3425 | 49.0 | 196 | 0.3704 | 1.0 |
0.3449 | 50.0 | 200 | 0.3702 | 1.0 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 3