N_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: 4.1743
• Accuracy: 0.5249

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
1.2626	1.0	534	1.1763	0.4769
0.969	2.0	1068	1.0746	0.5443
0.738	3.0	1602	1.2337	0.5276
0.5598	4.0	2136	1.4162	0.5312
0.3752	5.0	2670	1.7536	0.5176
0.2718	6.0	3204	1.9742	0.5285
0.2039	7.0	3738	2.2782	0.5299
0.153	8.0	4272	2.5862	0.5330
0.1248	9.0	4806	2.7811	0.5371
0.1044	10.0	5340	3.1522	0.5231
0.0912	11.0	5874	3.3783	0.5145
0.0733	12.0	6408	3.5910	0.5281
0.0447	13.0	6942	3.6989	0.5240
0.0345	14.0	7476	3.8076	0.5267
0.0321	15.0	8010	3.9878	0.5154
0.0205	16.0	8544	4.0187	0.5208
0.0123	17.0	9078	4.0994	0.5208
0.0127	18.0	9612	4.1125	0.5321
0.0092	19.0	10146	4.1869	0.5217
0.0094	20.0	10680	4.1743	0.5249

Framework versions

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