mrpc_normal

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2387
• Accuracy: 0.8446

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5046	1.0	255	0.4014	0.8104
0.2703	2.0	510	0.4722	0.8238
0.1435	3.0	765	0.6622	0.8301
0.0773	4.0	1020	0.8342	0.8359
0.0606	5.0	1275	0.7360	0.8417
0.0407	6.0	1530	1.0387	0.8174
0.0257	7.0	1785	1.0302	0.8377
0.0128	8.0	2040	1.0569	0.8383
0.0143	9.0	2295	1.0179	0.8371
0.0135	10.0	2550	1.0698	0.8417
0.0085	11.0	2805	1.0444	0.8493
0.0066	12.0	3060	1.1725	0.84
0.0022	13.0	3315	1.1954	0.8412
0.003	14.0	3570	1.2206	0.8464
0.0022	15.0	3825	1.2387	0.8446

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2