STS-conventional-Fine-Tuning-Capstone-roberta-base-filtered-275

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

• Loss: 2.2004
• Accuracy: 0.7341

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• 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
No log	1.0	113	0.8470	0.6966
No log	2.0	226	1.0989	0.6573
No log	3.0	339	1.0727	0.6854
No log	4.0	452	1.2050	0.6498
0.2702	5.0	565	1.4546	0.6873
0.2702	6.0	678	1.4088	0.7228
0.2702	7.0	791	1.7711	0.7154
0.2702	8.0	904	1.8350	0.7303
0.1029	9.0	1017	1.8810	0.7303
0.1029	10.0	1130	1.9587	0.7210
0.1029	11.0	1243	2.1934	0.7154
0.1029	12.0	1356	2.1518	0.7322
0.1029	13.0	1469	2.1331	0.7434
0.0393	14.0	1582	2.1823	0.7397
0.0393	15.0	1695	2.2004	0.7341

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2