N_roberta_sst5_padding0model

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

• Loss: 3.9677
• 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
1.231	1.0	534	1.1949	0.4588
0.9709	2.0	1068	1.0009	0.5643
0.826	3.0	1602	1.0707	0.5548
0.6894	4.0	2136	1.1785	0.5566
0.5545	5.0	2670	1.2947	0.5566
0.4264	6.0	3204	1.4867	0.5448
0.3507	7.0	3738	1.6741	0.5584
0.2579	8.0	4272	1.8731	0.5643
0.2188	9.0	4806	2.2232	0.5566
0.1903	10.0	5340	2.5289	0.5520
0.1664	11.0	5874	2.9037	0.5633
0.1563	12.0	6408	3.1956	0.5620
0.1152	13.0	6942	3.3490	0.5538
0.1064	14.0	7476	3.4903	0.5561
0.0774	15.0	8010	3.6072	0.5602
0.0585	16.0	8544	3.7163	0.5584
0.048	17.0	9078	3.8144	0.5633
0.0418	18.0	9612	3.9285	0.5584
0.033	19.0	10146	3.9853	0.5575
0.0355	20.0	10680	3.9677	0.5593

Framework versions

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