mobilebert_sa_GLUE_Experiment_data_aug_stsb
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.8342
- Pearson: 0.1765
- Spearmanr: 0.1800
- Combined Score: 0.1782
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 | Pearson | Spearmanr | Combined Score |
---|---|---|---|---|---|---|
1.0254 | 1.0 | 2518 | 2.8776 | 0.1575 | 0.1742 | 0.1659 |
0.5854 | 2.0 | 5036 | 3.1464 | 0.1591 | 0.1679 | 0.1635 |
0.4255 | 3.0 | 7554 | 2.8342 | 0.1765 | 0.1800 | 0.1782 |
0.2765 | 4.0 | 10072 | 2.8524 | 0.1815 | 0.1838 | 0.1827 |
0.1862 | 5.0 | 12590 | 2.9184 | 0.1736 | 0.1768 | 0.1752 |
0.1339 | 6.0 | 15108 | 2.9817 | 0.1688 | 0.1728 | 0.1708 |
0.1029 | 7.0 | 17626 | 2.9702 | 0.1618 | 0.1643 | 0.1631 |
0.0806 | 8.0 | 20144 | 3.0033 | 0.1588 | 0.1624 | 0.1606 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2
- Downloads last month
- 2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.