bert_sst5_padding20model

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: 4.1200
• Accuracy: 0.5303

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
1.3044	1.0	534	1.2199	0.4462
1.0097	2.0	1068	1.0752	0.5367
0.789	3.0	1602	1.1757	0.5312
0.6158	4.0	2136	1.3412	0.5281
0.4431	5.0	2670	1.7108	0.5050
0.3048	6.0	3204	1.8904	0.5199
0.2302	7.0	3738	2.1967	0.5195
0.1817	8.0	4272	2.4892	0.5167
0.1529	9.0	4806	2.6371	0.5335
0.1182	10.0	5340	3.0192	0.5231
0.1016	11.0	5874	3.2814	0.5186
0.0826	12.0	6408	3.4722	0.5258
0.0616	13.0	6942	3.6937	0.5199
0.0545	14.0	7476	3.7327	0.5231
0.0326	15.0	8010	3.8604	0.5235
0.022	16.0	8544	3.9752	0.5244
0.0197	17.0	9078	4.0806	0.5222
0.0125	18.0	9612	4.0842	0.5317
0.0115	19.0	10146	4.1118	0.5294
0.011	20.0	10680	4.1200	0.5303

Framework versions

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