roberta_sst2_padding80model

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5696
• Accuracy: 0.9385

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: 2e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	433	0.1681	0.9368
0.332	2.0	866	0.3784	0.9099
0.1989	3.0	1299	0.2267	0.9478
0.1089	4.0	1732	0.3591	0.9341
0.0727	5.0	2165	0.3262	0.9429
0.0381	6.0	2598	0.3532	0.9385
0.0307	7.0	3031	0.4867	0.9341
0.0307	8.0	3464	0.4819	0.9352
0.0201	9.0	3897	0.5591	0.9220
0.0249	10.0	4330	0.5679	0.9319
0.0142	11.0	4763	0.4514	0.9379
0.0199	12.0	5196	0.4652	0.9374
0.0095	13.0	5629	0.5044	0.9451
0.0111	14.0	6062	0.4807	0.9407
0.0111	15.0	6495	0.4752	0.9407
0.009	16.0	6928	0.5267	0.9374
0.0046	17.0	7361	0.5300	0.9396
0.0071	18.0	7794	0.5414	0.9412
0.0006	19.0	8227	0.5648	0.9407
0.0024	20.0	8660	0.5696	0.9385

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1
• Datasets 2.12.0
• Tokenizers 0.13.3