roberta_sst2_padding60model

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

• Loss: 0.5143
• Accuracy: 0.9445

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	0.2571	0.9083
0.3356	2.0	866	0.1977	0.9407
0.1872	3.0	1299	0.2530	0.9396
0.1195	4.0	1732	0.2908	0.9434
0.0754	5.0	2165	0.3693	0.9412
0.043	6.0	2598	0.3650	0.9379
0.0317	7.0	3031	0.4696	0.9390
0.0317	8.0	3464	0.4233	0.9434
0.0277	9.0	3897	0.4518	0.9385
0.0248	10.0	4330	0.4169	0.9418
0.0156	11.0	4763	0.5964	0.9325
0.0116	12.0	5196	0.4787	0.9412
0.0093	13.0	5629	0.4558	0.9456
0.0129	14.0	6062	0.5166	0.9330
0.0129	15.0	6495	0.5489	0.9407
0.0068	16.0	6928	0.5002	0.9440
0.0059	17.0	7361	0.5227	0.9445
0.002	18.0	7794	0.5155	0.9462
0.0032	19.0	8227	0.5027	0.9473
0.0008	20.0	8660	0.5143	0.9445

Framework versions

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