metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: SER_model_xapiens
results: []
SER_model_xapiens
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: 1.9265
- Accuracy: 0.3565
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- 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 |
---|---|---|---|---|
1.4373 | 1.0 | 12 | 1.8861 | 0.3578 |
1.501 | 2.0 | 24 | 1.8790 | 0.3735 |
1.5003 | 3.0 | 36 | 1.8961 | 0.3696 |
1.423 | 4.0 | 48 | 1.8633 | 0.3591 |
1.4632 | 5.0 | 60 | 1.8994 | 0.3526 |
1.4364 | 6.0 | 72 | 1.9166 | 0.3539 |
1.4105 | 7.0 | 84 | 1.9423 | 0.3499 |
1.3816 | 8.0 | 96 | 1.9297 | 0.3434 |
1.4097 | 9.0 | 108 | 1.9315 | 0.3578 |
1.3785 | 10.0 | 120 | 1.9265 | 0.3565 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.3.0
- Datasets 2.19.2.dev0
- Tokenizers 0.19.1