bert_sst5_padding30model

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: 4.1363
• Accuracy: 0.5217

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
1.3309	1.0	534	1.1877	0.4566
1.0031	2.0	1068	1.0797	0.5439
0.7657	3.0	1602	1.1662	0.5398
0.5876	4.0	2136	1.3887	0.5299
0.4158	5.0	2670	1.6423	0.5199
0.3096	6.0	3204	1.9704	0.5113
0.2247	7.0	3738	2.3219	0.5154
0.1668	8.0	4272	2.4251	0.5308
0.1559	9.0	4806	2.8634	0.5204
0.1096	10.0	5340	3.1091	0.5167
0.1087	11.0	5874	3.3826	0.5118
0.0866	12.0	6408	3.4875	0.5276
0.0674	13.0	6942	3.8364	0.5077
0.0556	14.0	7476	3.8687	0.5145
0.0403	15.0	8010	3.9204	0.5118
0.0228	16.0	8544	4.0594	0.5122
0.0229	17.0	9078	4.0215	0.5235
0.0155	18.0	9612	4.0793	0.5231
0.0088	19.0	10146	4.1288	0.5208
0.0127	20.0	10680	4.1363	0.5217

Framework versions

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