Sentiment-classfication-ROBERTA-model
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3282
- Accuracy: 0.9372
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.0985 | 0.14 | 100 | 1.0686 | 0.4783 |
0.8759 | 0.27 | 200 | 0.7441 | 0.6782 |
0.7197 | 0.41 | 300 | 0.8678 | 0.6422 |
0.7203 | 0.54 | 400 | 0.6434 | 0.7356 |
0.6408 | 0.68 | 500 | 0.6222 | 0.7675 |
0.6088 | 0.81 | 600 | 0.5258 | 0.8072 |
0.6058 | 0.95 | 700 | 0.5646 | 0.7977 |
0.4989 | 1.09 | 800 | 0.4470 | 0.8459 |
0.3946 | 1.22 | 900 | 0.4820 | 0.8333 |
0.4165 | 1.36 | 1000 | 0.3834 | 0.8595 |
0.3939 | 1.49 | 1100 | 0.4710 | 0.8323 |
0.3206 | 1.63 | 1200 | 0.3700 | 0.8758 |
0.3645 | 1.77 | 1300 | 0.3333 | 0.8917 |
0.3492 | 1.9 | 1400 | 0.3008 | 0.9016 |
0.2797 | 2.04 | 1500 | 0.3356 | 0.9012 |
0.2036 | 2.17 | 1600 | 0.3982 | 0.9026 |
0.225 | 2.31 | 1700 | 0.3478 | 0.9060 |
0.1952 | 2.44 | 1800 | 0.3572 | 0.9134 |
0.1919 | 2.58 | 1900 | 0.3718 | 0.9114 |
0.2243 | 2.72 | 2000 | 0.3335 | 0.9219 |
0.1779 | 2.85 | 2100 | 0.3221 | 0.9277 |
0.2047 | 2.99 | 2200 | 0.3232 | 0.9338 |
0.1023 | 3.12 | 2300 | 0.3767 | 0.9304 |
0.1253 | 3.26 | 2400 | 0.3686 | 0.9314 |
0.1157 | 3.39 | 2500 | 0.3282 | 0.9372 |
0.0995 | 3.53 | 2600 | 0.3662 | 0.9308 |
0.1176 | 3.67 | 2700 | 0.3182 | 0.9331 |
0.0982 | 3.8 | 2800 | 0.3224 | 0.9348 |
0.0674 | 3.94 | 2900 | 0.3222 | 0.9355 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 11