bert_sst5_padding50model

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: 4.1434
• Accuracy: 0.5357

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	1.2481	0.4520
0.9866	2.0	1068	1.0774	0.5416
0.7639	3.0	1602	1.1460	0.5294
0.5824	4.0	2136	1.4723	0.5136
0.4104	5.0	2670	1.6064	0.5294
0.3146	6.0	3204	1.8671	0.5231
0.2298	7.0	3738	2.1795	0.5181
0.1771	8.0	4272	2.3543	0.5321
0.1429	9.0	4806	2.7375	0.5222
0.1153	10.0	5340	3.1331	0.5339
0.0981	11.0	5874	3.2020	0.5403
0.0764	12.0	6408	3.3910	0.5362
0.0637	13.0	6942	3.6035	0.5312
0.0418	14.0	7476	3.6734	0.5226
0.0352	15.0	8010	3.8465	0.5271
0.0267	16.0	8544	4.0115	0.5240
0.0166	17.0	9078	3.9827	0.5290
0.0125	18.0	9612	4.0704	0.5335
0.0078	19.0	10146	4.1352	0.5290
0.0062	20.0	10680	4.1434	0.5357

Framework versions

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