roberta_sst5_padding50model

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

• Loss: 3.9057
• Accuracy: 0.5593

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
1.38	1.0	534	1.2160	0.4516
1.0421	2.0	1068	1.0457	0.5520
0.8678	3.0	1602	1.0607	0.5570
0.7334	4.0	2136	1.0964	0.5701
0.5777	5.0	2670	1.2672	0.5665
0.4545	6.0	3204	1.4068	0.5647
0.3607	7.0	3738	1.6500	0.5643
0.2646	8.0	4272	1.8139	0.5638
0.2394	9.0	4806	2.0395	0.5561
0.1846	10.0	5340	2.4762	0.5602
0.1789	11.0	5874	2.8201	0.5471
0.1484	12.0	6408	2.9387	0.5593
0.1307	13.0	6942	3.2414	0.5629
0.1004	14.0	7476	3.3631	0.5674
0.0705	15.0	8010	3.5533	0.5584
0.0655	16.0	8544	3.5689	0.5724
0.0482	17.0	9078	3.7737	0.5593
0.0484	18.0	9612	3.8848	0.5557
0.0414	19.0	10146	3.8977	0.5588
0.0284	20.0	10680	3.9057	0.5593

Framework versions

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