STS-conventional-Fine-Tuning-Capstone-roberta-base-filtered-160

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: 2.2762
• Accuracy: 0.7247

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	113	0.7940	0.6610
No log	2.0	226	0.7463	0.6929
No log	3.0	339	0.9240	0.7041
No log	4.0	452	0.9070	0.6629
0.5167	5.0	565	1.1376	0.7022
0.5167	6.0	678	1.2043	0.7022
0.5167	7.0	791	1.3083	0.7228
0.5167	8.0	904	1.5205	0.7154
0.1626	9.0	1017	1.5875	0.7154
0.1626	10.0	1130	1.8172	0.7041
0.1626	11.0	1243	1.9300	0.7154
0.1626	12.0	1356	1.8632	0.7247
0.1626	13.0	1469	2.0908	0.7135
0.0655	14.0	1582	2.0766	0.7191
0.0655	15.0	1695	2.2582	0.7135
0.0655	16.0	1808	2.2743	0.7154
0.0655	17.0	1921	2.2310	0.7228
0.0237	18.0	2034	2.2574	0.7285
0.0237	19.0	2147	2.2768	0.7266
0.0237	20.0	2260	2.2762	0.7247

Framework versions

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