sentiment-lora-r2a0d0.15-0

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: 0.3672
• Accuracy: 0.8321
• Precision: 0.7961
• Recall: 0.8087
• F1: 0.8018

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
0.563	1.0	122	0.5138	0.7243	0.6636	0.6549	0.6586
0.509	2.0	244	0.5057	0.7168	0.6763	0.6996	0.6820
0.4924	3.0	366	0.4708	0.7393	0.6877	0.6931	0.6901
0.468	4.0	488	0.4379	0.7845	0.7412	0.7200	0.7286
0.4495	5.0	610	0.4466	0.7594	0.7233	0.7548	0.7313
0.4334	6.0	732	0.4041	0.8271	0.7927	0.7851	0.7887
0.415	7.0	854	0.4057	0.7995	0.7590	0.7756	0.7660
0.3974	8.0	976	0.3852	0.8321	0.7982	0.7937	0.7959
0.3849	9.0	1098	0.3829	0.8246	0.7880	0.7909	0.7894
0.3771	10.0	1220	0.3786	0.8396	0.8065	0.8065	0.8065
0.3633	11.0	1342	0.3843	0.8296	0.7931	0.8069	0.7993
0.3591	12.0	1464	0.3833	0.8296	0.7931	0.8069	0.7993
0.354	13.0	1586	0.3705	0.8396	0.8065	0.8065	0.8065
0.3451	14.0	1708	0.3709	0.8371	0.8028	0.8072	0.8049
0.3403	15.0	1830	0.3733	0.8321	0.7960	0.8112	0.8027
0.3282	16.0	1952	0.3715	0.8346	0.7988	0.8155	0.8061
0.3286	17.0	2074	0.3664	0.8321	0.7965	0.8037	0.7999
0.3348	18.0	2196	0.3670	0.8271	0.7904	0.8001	0.7949
0.325	19.0	2318	0.3669	0.8321	0.7961	0.8087	0.8018
0.3266	20.0	2440	0.3672	0.8321	0.7961	0.8087	0.8018

Framework versions

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