sentiment-base-3

This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8030
• Accuracy: 0.9023
• Precision: 0.8875
• Recall: 0.8733
• F1: 0.8799

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: 30
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.4092	1.0	122	0.3457	0.8521	0.8930	0.7554	0.7892
0.2282	2.0	244	0.2584	0.8922	0.8749	0.8612	0.8676
0.138	3.0	366	0.4417	0.8797	0.8825	0.8199	0.8430
0.0837	4.0	488	0.4037	0.9023	0.8893	0.8708	0.8793
0.0426	5.0	610	0.5462	0.9048	0.8806	0.8951	0.8873
0.0502	6.0	732	0.5626	0.8897	0.8618	0.8820	0.8707
0.0242	7.0	854	0.6241	0.9073	0.8977	0.8744	0.8849
0.0217	8.0	976	0.7096	0.8872	0.8579	0.8852	0.8692
0.0229	9.0	1098	0.6115	0.9123	0.8910	0.9004	0.8955
0.0109	10.0	1220	0.7575	0.8972	0.8796	0.8698	0.8745
0.0068	11.0	1342	0.7537	0.9073	0.8938	0.8794	0.8861
0.0131	12.0	1464	0.7247	0.8972	0.8732	0.8823	0.8776
0.0101	13.0	1586	0.7928	0.8972	0.8754	0.8773	0.8764
0.0061	14.0	1708	0.7849	0.9073	0.8875	0.8894	0.8884
0.0135	15.0	1830	0.7816	0.8972	0.8830	0.8648	0.8731
0.0081	16.0	1952	0.7727	0.8972	0.8767	0.8748	0.8757
0.0027	17.0	2074	0.8128	0.8972	0.8754	0.8773	0.8764
0.0041	18.0	2196	0.8081	0.9023	0.8828	0.8808	0.8818
0.0018	19.0	2318	0.8039	0.9023	0.8893	0.8708	0.8793
0.0025	20.0	2440	0.8030	0.9023	0.8875	0.8733	0.8799

Framework versions

• Transformers 4.39.3
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.15.2