|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: wav2vec-best-CREMA-sentiment-analysis |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# wav2vec-best-CREMA-sentiment-analysis |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Top2 Accuracy: 0.8940 |
|
- Loss: 0.8287 |
|
- Accuracy: 0.7074 |
|
|
|
## 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: 1e-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: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Top2 Accuracy | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:-------------:|:---------------:|:--------:| |
|
| 1.7824 | 0.98 | 43 | 0.4982 | 1.7749 | 0.2482 | |
|
| 1.7115 | 1.99 | 87 | 0.5466 | 1.6566 | 0.3638 | |
|
| 1.5255 | 2.99 | 131 | 0.6604 | 1.5017 | 0.4418 | |
|
| 1.3716 | 4.0 | 175 | 0.7679 | 1.3359 | 0.5636 | |
|
| 1.2436 | 4.98 | 218 | 0.8271 | 1.1862 | 0.6407 | |
|
| 1.1366 | 5.99 | 262 | 0.8315 | 1.1223 | 0.6595 | |
|
| 1.0322 | 6.99 | 306 | 0.8593 | 1.0422 | 0.6747 | |
|
| 0.9668 | 8.0 | 350 | 0.8907 | 0.9335 | 0.7222 | |
|
| 0.8932 | 8.98 | 393 | 0.8943 | 0.9093 | 0.7231 | |
|
| 0.8431 | 9.99 | 437 | 0.8692 | 0.9163 | 0.7115 | |
|
| 0.8047 | 10.99 | 481 | 0.8996 | 0.8488 | 0.7375 | |
|
| 0.7444 | 12.0 | 525 | 0.8898 | 0.8611 | 0.7204 | |
|
| 0.6921 | 12.98 | 568 | 0.8916 | 0.8399 | 0.7258 | |
|
| 0.6973 | 13.99 | 612 | 0.8844 | 0.8425 | 0.7231 | |
|
| 0.632 | 14.99 | 656 | 0.8880 | 0.8308 | 0.7249 | |
|
| 0.6275 | 16.0 | 700 | 0.8862 | 0.8400 | 0.7177 | |
|
| 0.6153 | 16.98 | 743 | 0.8934 | 0.8266 | 0.7330 | |
|
| 0.5597 | 17.99 | 787 | 0.8934 | 0.8157 | 0.7357 | |
|
| 0.5658 | 18.99 | 831 | 0.8862 | 0.8015 | 0.7446 | |
|
| 0.54 | 20.0 | 875 | 0.8943 | 0.8368 | 0.7258 | |
|
| 0.5301 | 20.98 | 918 | 0.9023 | 0.8095 | 0.7321 | |
|
| 0.5262 | 21.99 | 962 | 0.8817 | 0.8521 | 0.7168 | |
|
| 0.4754 | 22.99 | 1006 | 0.8987 | 0.8003 | 0.7428 | |
|
| 0.4753 | 24.0 | 1050 | 0.8952 | 0.7988 | 0.7410 | |
|
| 0.455 | 24.98 | 1093 | 0.8952 | 0.7902 | 0.7419 | |
|
| 0.4574 | 25.99 | 1137 | 0.8871 | 0.8030 | 0.7366 | |
|
| 0.4618 | 26.99 | 1181 | 0.8970 | 0.8051 | 0.7294 | |
|
| 0.4222 | 28.0 | 1225 | 0.8925 | 0.8108 | 0.7267 | |
|
| 0.4301 | 28.98 | 1268 | 0.8934 | 0.8066 | 0.7339 | |
|
| 0.4147 | 29.49 | 1290 | 0.8916 | 0.8072 | 0.7357 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
