mobilebert_add_GLUE_Experiment_sst2_128
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE SST2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4543
- Accuracy: 0.7982
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: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6677 | 1.0 | 527 | 0.6771 | 0.5757 |
| 0.5966 | 2.0 | 1054 | 0.7135 | 0.5424 |
| 0.5714 | 3.0 | 1581 | 0.7271 | 0.5550 |
| 0.5573 | 4.0 | 2108 | 0.6892 | 0.5619 |
| 0.501 | 5.0 | 2635 | 0.4546 | 0.7798 |
| 0.2856 | 6.0 | 3162 | 0.4613 | 0.8050 |
| 0.2288 | 7.0 | 3689 | 0.4543 | 0.7982 |
| 0.2027 | 8.0 | 4216 | 0.4662 | 0.7993 |
| 0.1883 | 9.0 | 4743 | 0.5168 | 0.8039 |
| 0.1779 | 10.0 | 5270 | 0.5748 | 0.7856 |
| 0.1691 | 11.0 | 5797 | 0.5196 | 0.8028 |
| 0.1596 | 12.0 | 6324 | 0.5943 | 0.7947 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2