<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
mobilebert_add_GLUE_Experiment_qqp
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.5008
- Accuracy: 0.7600
- F1: 0.6402
- Combined Score: 0.7001
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 | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.6505 | 1.0 | 2843 | 0.6498 | 0.6321 | 0.0012 | 0.3166 |
| 0.6474 | 2.0 | 5686 | 0.6484 | 0.6321 | 0.0012 | 0.3166 |
| 0.646 | 3.0 | 8529 | 0.6479 | 0.6322 | 0.0024 | 0.3173 |
| 0.5481 | 4.0 | 11372 | 0.5140 | 0.7486 | 0.6247 | 0.6867 |
| 0.4934 | 5.0 | 14215 | 0.5086 | 0.7529 | 0.6548 | 0.7039 |
| 0.4794 | 6.0 | 17058 | 0.5044 | 0.7575 | 0.6527 | 0.7051 |
| 0.4708 | 7.0 | 19901 | 0.5008 | 0.7600 | 0.6402 | 0.7001 |
| 0.4652 | 8.0 | 22744 | 0.5010 | 0.7619 | 0.6384 | 0.7001 |
| 0.4604 | 9.0 | 25587 | 0.5014 | 0.7614 | 0.6489 | 0.7052 |
| 0.4562 | 10.0 | 28430 | 0.5057 | 0.7600 | 0.6617 | 0.7108 |
| 0.452 | 11.0 | 31273 | 0.5102 | 0.7620 | 0.6364 | 0.6992 |
| 0.4476 | 12.0 | 34116 | 0.5302 | 0.7622 | 0.6619 | 0.7121 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2
