N_roberta_sst5_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: 1.3783
• Accuracy: 0.4276

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
1.5765	1.0	534	1.5921	0.2308
1.5735	2.0	1068	1.5832	0.2308
1.572	3.0	1602	1.5875	0.2308
1.5702	4.0	2136	1.5799	0.2308
1.5752	5.0	2670	1.5783	0.2864
1.574	6.0	3204	1.5806	0.2308
1.5722	7.0	3738	1.5792	0.2308
1.5711	8.0	4272	1.5805	0.2308
1.5703	9.0	4806	1.5822	0.2308
1.5676	10.0	5340	1.5820	0.2308
1.5733	11.0	5874	1.5823	0.2308
1.5716	12.0	6408	1.5820	0.2308
1.5698	13.0	6942	1.5816	0.2308
1.5708	14.0	7476	1.5834	0.2308
1.5701	15.0	8010	1.5816	0.2308
1.5699	16.0	8544	1.5806	0.2308
1.5704	17.0	9078	1.5808	0.2308
1.5168	18.0	9612	1.4289	0.4167
1.4215	19.0	10146	1.3884	0.4231
1.3879	20.0	10680	1.3783	0.4276

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3