roberta_sst2_padding70model

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

• Loss: 0.5675
• Accuracy: 0.9412

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	0.1869	0.9308
0.3423	2.0	866	0.4539	0.9039
0.2033	3.0	1299	0.2940	0.9379
0.1087	4.0	1732	0.3001	0.9396
0.0742	5.0	2165	0.3625	0.9379
0.0511	6.0	2598	0.4227	0.9407
0.028	7.0	3031	0.4785	0.9352
0.028	8.0	3464	0.3861	0.9412
0.023	9.0	3897	0.4512	0.9407
0.0201	10.0	4330	0.5132	0.9368
0.0123	11.0	4763	0.4452	0.9423
0.0115	12.0	5196	0.4293	0.9500
0.0071	13.0	5629	0.7280	0.9182
0.0186	14.0	6062	0.5646	0.9368
0.0186	15.0	6495	0.5034	0.9434
0.0049	16.0	6928	0.5029	0.9418
0.0078	17.0	7361	0.4935	0.9456
0.0039	18.0	7794	0.5453	0.9418
0.0027	19.0	8227	0.5793	0.9385
0.0006	20.0	8660	0.5675	0.9412

Framework versions

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