metadata
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: sentiment_deberta
results: []
sentiment_deberta
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7123
- Accuracy: 0.6938
- F1: 0.6401
- Precision: 0.6262
- Recall: 0.6854
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 1.087 | 1.0 | 47 | 1.1008 | 0.2551 | 0.3042 | 0.4734 | 0.4956 |
| 0.9933 | 2.0 | 94 | 0.9692 | 0.5545 | 0.5098 | 0.5126 | 0.5496 |
| 0.8709 | 3.0 | 141 | 0.9352 | 0.5003 | 0.5003 | 0.5301 | 0.5804 |
| 0.8444 | 4.0 | 188 | 0.8729 | 0.5874 | 0.5602 | 0.5671 | 0.6204 |
| 0.7833 | 5.0 | 235 | 0.9394 | 0.4778 | 0.4980 | 0.5643 | 0.6353 |
| 0.7003 | 6.0 | 282 | 0.7279 | 0.6834 | 0.6306 | 0.6150 | 0.6828 |
| 0.6383 | 7.0 | 329 | 0.7808 | 0.6390 | 0.6123 | 0.6073 | 0.7007 |
| 0.5996 | 8.0 | 376 | 0.7379 | 0.6802 | 0.6367 | 0.6231 | 0.6993 |
| 0.5514 | 9.0 | 423 | 0.7846 | 0.6745 | 0.6204 | 0.6015 | 0.6901 |
| 0.4837 | 10.0 | 470 | 0.7123 | 0.6938 | 0.6401 | 0.6262 | 0.6854 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1