bert-base-uncased-finetuned-paper

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: 0.2326
• Precision: 0.7612
• Recall: 0.7456
• F1: 0.7533
• Accuracy: 0.9684

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: 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	Precision	Recall	F1	Accuracy
No log	1.0	73	0.1917	0.6756	0.5175	0.5861	0.9484
No log	2.0	146	0.1402	0.7516	0.6988	0.7242	0.9678
No log	3.0	219	0.1747	0.7397	0.6813	0.7093	0.9659
No log	4.0	292	0.1627	0.6797	0.7632	0.7190	0.9633
No log	5.0	365	0.1720	0.7005	0.7456	0.7224	0.9661
No log	6.0	438	0.2029	0.7515	0.7339	0.7426	0.9688
0.0876	7.0	511	0.1928	0.7415	0.7632	0.7522	0.9700
0.0876	8.0	584	0.2016	0.7579	0.7690	0.7634	0.9708
0.0876	9.0	657	0.2051	0.7371	0.7544	0.7457	0.9684
0.0876	10.0	730	0.2153	0.7477	0.7281	0.7378	0.9693
0.0876	11.0	803	0.2284	0.7626	0.7515	0.7570	0.9693
0.0876	12.0	876	0.2223	0.7139	0.7515	0.7322	0.9682
0.0876	13.0	949	0.2274	0.7471	0.7515	0.7493	0.9690
0.0022	14.0	1022	0.2321	0.7695	0.7515	0.7604	0.9695
0.0022	15.0	1095	0.2367	0.7590	0.7368	0.7478	0.9690
0.0022	16.0	1168	0.2327	0.7612	0.7456	0.7533	0.9695
0.0022	17.0	1241	0.2367	0.7704	0.7456	0.7578	0.9690
0.0022	18.0	1314	0.2309	0.7529	0.7485	0.7507	0.9691
0.0022	19.0	1387	0.2358	0.7711	0.7485	0.7596	0.9686
0.0022	20.0	1460	0.2326	0.7612	0.7456	0.7533	0.9684

Framework versions

• Transformers 4.27.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2