results

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: 0.5524
• Precision: 0.6970
• Recall: 0.6661
• F1: 0.6798
• Accuracy: 0.9081

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	Precision	Recall	F1	Accuracy
No log	1.0	250	0.2537	0.7645	0.5606	0.5850	0.9188
0.2695	2.0	500	0.2528	0.7559	0.6099	0.6464	0.9205
0.2695	3.0	750	0.2524	0.7359	0.6409	0.6729	0.9183
0.2201	4.0	1000	0.2660	0.7015	0.6596	0.6773	0.9099
0.2201	5.0	1250	0.2926	0.6924	0.6821	0.6870	0.9053
0.1629	6.0	1500	0.3055	0.6904	0.6848	0.6876	0.9042
0.1629	7.0	1750	0.3332	0.7037	0.6532	0.6736	0.9109
0.1144	8.0	2000	0.3661	0.6870	0.6759	0.6812	0.9038
0.1144	9.0	2250	0.3670	0.6950	0.6597	0.6750	0.9079
0.081	10.0	2500	0.4031	0.6969	0.6588	0.6751	0.9086
0.081	11.0	2750	0.4176	0.6883	0.6734	0.6804	0.9045
0.0611	12.0	3000	0.4531	0.7003	0.6552	0.6739	0.9098
0.0611	13.0	3250	0.4733	0.6970	0.6600	0.6758	0.9085
0.0476	14.0	3500	0.4815	0.6997	0.6533	0.6724	0.9098
0.0476	15.0	3750	0.5058	0.6977	0.6580	0.6748	0.9089
0.039	16.0	4000	0.5027	0.7011	0.6646	0.6804	0.9095
0.039	17.0	4250	0.5196	0.6993	0.6635	0.6790	0.9090
0.0309	18.0	4500	0.5462	0.6986	0.6687	0.6819	0.9085
0.0309	19.0	4750	0.5406	0.6939	0.6684	0.6799	0.9069
0.0273	20.0	5000	0.5524	0.6970	0.6661	0.6798	0.9081

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1