N_roberta_sst5_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: 1.5808
• Accuracy: 0.2308

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
1.5766	1.0	534	1.5965	0.2308
1.5738	2.0	1068	1.5814	0.2308
1.5746	3.0	1602	1.5872	0.2308
1.5693	4.0	2136	1.5791	0.2308
1.5735	5.0	2670	1.5786	0.2864
1.5731	6.0	3204	1.5841	0.2308
1.5713	7.0	3738	1.5772	0.2308
1.571	8.0	4272	1.5803	0.2308
1.5723	9.0	4806	1.5799	0.2308
1.5682	10.0	5340	1.5815	0.2308
1.5716	11.0	5874	1.5797	0.2308
1.5705	12.0	6408	1.5801	0.2864
1.5695	13.0	6942	1.5837	0.2308
1.5709	14.0	7476	1.5809	0.2308
1.5691	15.0	8010	1.5828	0.2308
1.57	16.0	8544	1.5824	0.2308
1.5688	17.0	9078	1.5814	0.2308
1.5707	18.0	9612	1.5809	0.2308
1.5685	19.0	10146	1.5809	0.2308
1.5691	20.0	10680	1.5808	0.2308

Framework versions

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