N_roberta_sst5_padding20model

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

• Loss: 3.8874
• Accuracy: 0.5665

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
1.2989	1.0	534	1.1466	0.5077
1.0029	2.0	1068	0.9966	0.5647
0.8473	3.0	1602	1.0603	0.5674
0.7264	4.0	2136	1.1194	0.5647
0.5747	5.0	2670	1.2826	0.5493
0.4677	6.0	3204	1.4385	0.5638
0.3581	7.0	3738	1.6142	0.5552
0.2666	8.0	4272	1.8359	0.5611
0.2181	9.0	4806	2.1608	0.5471
0.1815	10.0	5340	2.3635	0.5543
0.1629	11.0	5874	2.8433	0.5489
0.1448	12.0	6408	3.1203	0.5525
0.107	13.0	6942	3.1453	0.5710
0.1121	14.0	7476	3.5024	0.5620
0.0853	15.0	8010	3.4952	0.5688
0.0669	16.0	8544	3.6617	0.5588
0.0426	17.0	9078	3.7870	0.5647
0.0445	18.0	9612	3.9028	0.5584
0.0377	19.0	10146	3.8831	0.5674
0.0228	20.0	10680	3.8874	0.5665

Framework versions

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