bsc_ai_thesis_torgo_model-1
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3532
- Accuracy: 0.8625
- Precision: 0.8349
- Recall: 0.9055
- F1: 0.8687
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-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|---|---|---|
0.6855 | 0.96 | 12 | 0.6603 | 0.6225 | 0.5772 | 0.9303 | 0.7124 |
0.5875 | 2.0 | 25 | 0.5249 | 0.785 | 0.7533 | 0.8507 | 0.7991 |
0.4858 | 2.96 | 37 | 0.5584 | 0.7575 | 0.6940 | 0.9254 | 0.7932 |
0.3951 | 4.0 | 50 | 0.5366 | 0.785 | 0.7220 | 0.9303 | 0.8130 |
0.3962 | 4.96 | 62 | 0.4707 | 0.805 | 0.7450 | 0.9303 | 0.8274 |
0.3069 | 6.0 | 75 | 0.4032 | 0.8325 | 0.8190 | 0.8557 | 0.8370 |
0.2973 | 6.96 | 87 | 0.3753 | 0.855 | 0.8593 | 0.8507 | 0.855 |
0.2585 | 8.0 | 100 | 0.3719 | 0.8625 | 0.8259 | 0.9204 | 0.8706 |
0.2365 | 8.96 | 112 | 0.3503 | 0.855 | 0.8357 | 0.8856 | 0.8599 |
0.2244 | 9.6 | 120 | 0.3532 | 0.8625 | 0.8349 | 0.9055 | 0.8687 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
- Downloads last month
- 9
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.