small-vanilla-target-imdb
This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the imdb dataset. It achieves the following results on the evaluation set:
- Loss: 0.7710
- Accuracy: 0.8146
- F1: 0.8978
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.3417 | 0.64 | 500 | 0.1678 | 0.9286 | 0.9630 |
| 0.2401 | 1.28 | 1000 | 0.1262 | 0.9525 | 0.9757 |
| 0.1907 | 1.92 | 1500 | 0.2724 | 0.8963 | 0.9453 |
| 0.1397 | 2.56 | 2000 | 0.2378 | 0.9247 | 0.9609 |
| 0.11 | 3.2 | 2500 | 0.7710 | 0.8146 | 0.8978 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.13.2