584_test1_1

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9307
• Accuracy: 0.7354

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	241	1.0720	0.5229
No log	2.0	482	0.7864	0.7021
1.0917	3.0	723	0.8088	0.6917
1.0917	4.0	964	0.7681	0.725
0.5113	5.0	1205	1.1190	0.7208
0.5113	6.0	1446	1.1611	0.7188
0.1988	7.0	1687	1.4737	0.7083
0.1988	8.0	1928	1.6724	0.7021
0.1019	9.0	2169	1.6648	0.7271
0.1019	10.0	2410	1.7081	0.7104
0.0621	11.0	2651	1.6761	0.7312
0.0621	12.0	2892	1.7188	0.7188
0.039	13.0	3133	1.7463	0.7417
0.039	14.0	3374	2.0024	0.725
0.0207	15.0	3615	1.9167	0.725
0.0207	16.0	3856	1.8888	0.7417
0.0131	17.0	4097	2.0032	0.7312
0.0131	18.0	4338	1.8946	0.7438
0.0087	19.0	4579	1.9148	0.7396
0.0087	20.0	4820	1.9307	0.7354

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1