N_bert_sst5_padding10model

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

• Loss: 4.3591
• Accuracy: 0.5222

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
1.2795	1.0	534	1.2715	0.4041
0.9732	2.0	1068	1.0766	0.5362
0.7556	3.0	1602	1.1870	0.5312
0.5656	4.0	2136	1.4145	0.5312
0.3935	5.0	2670	1.7418	0.5213
0.2852	6.0	3204	2.0376	0.5131
0.2193	7.0	3738	2.2655	0.5240
0.1606	8.0	4272	2.5819	0.5299
0.1355	9.0	4806	3.0238	0.5095
0.1165	10.0	5340	3.1689	0.5104
0.1143	11.0	5874	3.4023	0.5208
0.0756	12.0	6408	3.6903	0.5054
0.058	13.0	6942	3.8438	0.5176
0.0487	14.0	7476	3.8585	0.5195
0.0336	15.0	8010	3.9326	0.5195
0.0208	16.0	8544	4.1253	0.5217
0.0122	17.0	9078	4.2740	0.5186
0.0068	18.0	9612	4.2541	0.5226
0.0107	19.0	10146	4.3208	0.5213
0.0118	20.0	10680	4.3591	0.5222

Framework versions

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