<!-- 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-large-xlsr-53-english-pronunciation-evaluation-aod-oversampling
This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-english on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1381
- Accuracy: 0.7536
- F1: 0.7512
- Precision: 0.7510
- Recall: 0.7536
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
0.8559 | 1.0 | 313 | 0.8926 | 0.5948 | 0.5697 | 0.5782 | 0.5948 |
0.6647 | 2.0 | 626 | 0.8132 | 0.6528 | 0.6435 | 0.6478 | 0.6528 |
0.5562 | 3.0 | 939 | 0.7991 | 0.72 | 0.7197 | 0.7209 | 0.72 |
0.2166 | 4.0 | 1252 | 0.9808 | 0.7528 | 0.7515 | 0.7514 | 0.7528 |
0.0269 | 5.0 | 1565 | 1.1381 | 0.7536 | 0.7512 | 0.7510 | 0.7536 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3