pepppper/audio_cls

1---2library_name: transformers3license: apache-2.04base_model: facebook/wav2vec2-xls-r-300m5tags:6- generated_from_trainer7metrics:8- accuracy9model-index:10- name: audio_cls11  results: []12---13 14<!-- This model card has been generated automatically according to the information the Trainer had access to. You15should probably proofread and complete it, then remove this comment. -->16 17# audio_cls18 19This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.20It achieves the following results on the evaluation set:21- Loss: 2.654922- Accuracy: 0.033623 24## Model description25 26More information needed27 28## Intended uses & limitations29 30More information needed31 32## Training and evaluation data33 34More information needed35 36## Training procedure37 38### Training hyperparameters39 40The following hyperparameters were used during training:41- learning_rate: 0.000142- train_batch_size: 843- eval_batch_size: 844- seed: 4245- gradient_accumulation_steps: 246- total_train_batch_size: 1647- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments48- lr_scheduler_type: linear49- lr_scheduler_warmup_ratio: 0.150- num_epochs: 1051- mixed_precision_training: Native AMP52 53### Training results54 55| Training Loss | Epoch | Step | Validation Loss | Accuracy |56|:-------------:|:-----:|:----:|:---------------:|:--------:|57| 2.639         | 1.0   | 30   | 2.6473          | 0.0504   |58| 2.6399        | 2.0   | 60   | 2.6481          | 0.0672   |59| 2.6514        | 3.0   | 90   | 2.6458          | 0.0588   |60| 2.6384        | 4.0   | 120  | 2.6581          | 0.0336   |61| 2.6289        | 5.0   | 150  | 2.6557          | 0.0336   |62| 2.6308        | 6.0   | 180  | 2.6607          | 0.0252   |63| 2.643         | 7.0   | 210  | 2.6600          | 0.0336   |64| 2.648         | 8.0   | 240  | 2.6641          | 0.0336   |65| 2.6333        | 9.0   | 270  | 2.6596          | 0.0336   |66| 2.6289        | 10.0  | 300  | 2.6549          | 0.0336   |67 68 69### Framework versions70 71- Transformers 4.56.172- Pytorch 2.8.0+cu12673- Datasets 4.0.074- Tokenizers 0.22.075