sentiment-base-2

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: 0.8886
• Accuracy: 0.8922
• Precision: 0.8719
• Recall: 0.8662
• F1: 0.8690

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
0.3808	1.0	122	0.3794	0.8647	0.8737	0.7917	0.8186
0.221	2.0	244	0.2851	0.8722	0.8562	0.8271	0.8395
0.1363	3.0	366	0.3832	0.8947	0.8757	0.8680	0.8717
0.099	4.0	488	0.4968	0.8972	0.8869	0.8598	0.8717
0.0702	5.0	610	0.5205	0.8697	0.8503	0.8278	0.8377
0.0469	6.0	732	0.5740	0.8747	0.8552	0.8363	0.8448
0.0328	7.0	854	0.6012	0.8847	0.8581	0.8684	0.8629
0.0284	8.0	976	0.5403	0.8972	0.8812	0.8673	0.8738
0.019	9.0	1098	0.5909	0.8922	0.8657	0.8813	0.8728
0.016	10.0	1220	0.8931	0.8822	0.8694	0.8392	0.8521
0.0167	11.0	1342	0.6618	0.8972	0.8781	0.8723	0.8751
0.0168	12.0	1464	0.7513	0.9023	0.8842	0.8783	0.8812
0.0064	13.0	1586	0.7513	0.8997	0.8819	0.8741	0.8778
0.0078	14.0	1708	0.8152	0.8947	0.8789	0.8630	0.8704
0.0064	15.0	1830	0.7460	0.8997	0.8778	0.8816	0.8797
0.0055	16.0	1952	0.8232	0.8922	0.8734	0.8637	0.8683
0.006	17.0	2074	0.8421	0.8947	0.8757	0.8680	0.8717
0.0052	18.0	2196	0.8442	0.8872	0.8624	0.8677	0.8650
0.0035	19.0	2318	0.8841	0.8897	0.8682	0.8645	0.8663
0.0013	20.0	2440	0.8886	0.8922	0.8719	0.8662	0.8690

Framework versions

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