<!-- 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. -->
base-model-with-warmup-fairseq-V1
This model is a fine-tuned version of facebook/wav2vec2-base-960h on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.7695
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: 0.0001
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
15.5458 | 4.04 | 200 | 4.3505 |
3.9186 | 8.08 | 400 | 4.4677 |
3.7954 | 12.12 | 600 | 4.8597 |
3.8846 | 16.16 | 800 | 4.6803 |
3.7918 | 20.2 | 1000 | 4.6823 |
3.8406 | 24.24 | 1200 | 4.8027 |
3.7933 | 28.28 | 1400 | 4.7320 |
3.7851 | 32.32 | 1600 | 4.8768 |
3.7886 | 36.36 | 1800 | 4.7657 |
3.7775 | 40.4 | 2000 | 4.8205 |
3.8297 | 44.44 | 2200 | 4.7835 |
3.7944 | 48.48 | 2400 | 4.7695 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.2