mobilebert_sa_GLUE_Experiment_logit_kd_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: 1.1918
- Pearson: 0.1864
- Spearmanr: 0.1859
- Combined Score: 0.1862
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 |
---|---|---|---|---|---|---|
1.7465 | 1.0 | 45 | 1.2026 | 0.0588 | 0.0666 | 0.0627 |
1.079 | 2.0 | 90 | 1.4599 | 0.0595 | 0.0691 | 0.0643 |
1.0784 | 3.0 | 135 | 1.2063 | 0.0611 | 0.0707 | 0.0659 |
0.9943 | 4.0 | 180 | 1.3534 | 0.0730 | 0.0730 | 0.0730 |
0.9523 | 5.0 | 225 | 1.3943 | 0.1080 | 0.1010 | 0.1045 |
0.8379 | 6.0 | 270 | 1.1918 | 0.1864 | 0.1859 | 0.1862 |
0.7217 | 7.0 | 315 | 1.2542 | 0.2080 | 0.2144 | 0.2112 |
0.6304 | 8.0 | 360 | 1.2209 | 0.1920 | 0.1979 | 0.1950 |
0.5573 | 9.0 | 405 | 1.2925 | 0.1881 | 0.1814 | 0.1847 |
0.5048 | 10.0 | 450 | 1.3943 | 0.1731 | 0.1877 | 0.1804 |
0.4754 | 11.0 | 495 | 1.3058 | 0.1845 | 0.1817 | 0.1831 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2
- Downloads last month
- 7
Inference API (serverless) is not available, repository is disabled.
Dataset used to train gokuls/mobilebert_sa_GLUE_Experiment_logit_kd_stsb
Evaluation results
- Spearmanr on GLUE STSBvalidation set self-reported0.186