hBERTv2_new_no_pretrain_qnli
This model is a fine-tuned version of on the GLUE QNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.6386
- Accuracy: 0.6352
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: 4e-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 |
---|---|---|---|---|
0.6866 | 1.0 | 819 | 0.6770 | 0.5570 |
0.652 | 2.0 | 1638 | 0.6468 | 0.6264 |
0.5986 | 3.0 | 2457 | 0.6386 | 0.6352 |
0.5144 | 4.0 | 3276 | 0.6930 | 0.6590 |
0.4272 | 5.0 | 4095 | 0.7034 | 0.6548 |
0.3558 | 6.0 | 4914 | 0.8171 | 0.6637 |
0.2874 | 7.0 | 5733 | 0.9057 | 0.6601 |
0.2391 | 8.0 | 6552 | 1.0090 | 0.6445 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 5
Inference API (serverless) is not available, repository is disabled.
Dataset used to train gokuls/hBERTv2_new_no_pretrain_qnli
Evaluation results
- Accuracy on GLUE QNLIvalidation set self-reported0.635