bert-tiny-mlm-finetuned-imdb-finetuned-emotion
This model is a fine-tuned version of muhtasham/bert-tiny-mlm-finetuned-imdb on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.2401
- Accuracy: 0.9065
- F1: 0.9083
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
- 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 |
|---|---|---|---|---|---|
| 1.4581 | 2.0 | 500 | 1.0158 | 0.672 | 0.6025 |
| 0.8545 | 4.0 | 1000 | 0.6368 | 0.842 | 0.8344 |
| 0.5846 | 6.0 | 1500 | 0.4626 | 0.877 | 0.8745 |
| 0.422 | 8.0 | 2000 | 0.3520 | 0.8925 | 0.8917 |
| 0.3222 | 10.0 | 2500 | 0.2934 | 0.9035 | 0.9039 |
| 0.2695 | 12.0 | 3000 | 0.2573 | 0.9055 | 0.9056 |
| 0.2247 | 14.0 | 3500 | 0.2396 | 0.9115 | 0.9122 |
| 0.1986 | 16.0 | 4000 | 0.2311 | 0.9155 | 0.9154 |
| 0.1773 | 18.0 | 4500 | 0.2261 | 0.9125 | 0.9136 |
| 0.1554 | 20.0 | 5000 | 0.2309 | 0.914 | 0.9144 |
| 0.1407 | 22.0 | 5500 | 0.2401 | 0.9065 | 0.9083 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2