mobilebert_500exs_10timesteps_run1
This model is a fine-tuned version of google/mobilebert-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6723
- Accuracy: 0.58
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 16
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 32 | 0.7404 | 0.57 |
| No log | 2.0 | 64 | 0.7478 | 0.57 |
| No log | 3.0 | 96 | 0.7952 | 0.57 |
| No log | 4.0 | 128 | 0.6763 | 0.56 |
| No log | 5.0 | 160 | 0.6819 | 0.58 |
| No log | 6.0 | 192 | 0.6723 | 0.58 |
| No log | 7.0 | 224 | 0.6994 | 0.57 |
| No log | 8.0 | 256 | 0.8576 | 0.43 |
| No log | 9.0 | 288 | 0.7126 | 0.46 |
| No log | 10.0 | 320 | 0.7118 | 0.48 |
| No log | 11.0 | 352 | 0.7136 | 0.57 |
| No log | 12.0 | 384 | 0.6758 | 0.59 |
| No log | 13.0 | 416 | 0.6921 | 0.55 |
| No log | 14.0 | 448 | 0.6718 | 0.56 |
| No log | 15.0 | 480 | 0.6725 | 0.57 |
| 130076.376 | 16.0 | 512 | 0.6723 | 0.58 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu118
- Datasets 2.17.1
- Tokenizers 0.15.2