<!-- 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. -->
libri-alpha-0-Temp-1-mse
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.7799
- Wer: 0.2824
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: 2e-05
- train_batch_size: 28
- eval_batch_size: 28
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 56
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
23.5935 | 0.65 | 100 | 14.9492 | 0.2308 |
13.2557 | 1.31 | 200 | 5.0349 | 0.6077 |
9.9465 | 1.96 | 300 | 4.5289 | 0.5263 |
8.8716 | 2.61 | 400 | 4.1278 | 0.4377 |
8.2434 | 3.27 | 500 | 4.0123 | 0.4013 |
7.8993 | 3.92 | 600 | 3.8645 | 0.3500 |
7.554 | 4.58 | 700 | 4.0048 | 0.3276 |
7.3348 | 5.23 | 800 | 3.8652 | 0.3212 |
7.1906 | 5.88 | 900 | 3.7930 | 0.3101 |
6.9957 | 6.54 | 1000 | 3.4559 | 0.2958 |
6.9153 | 7.19 | 1100 | 3.7799 | 0.2824 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.11.0