sentiment-lora-r2a2d0.15-1

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: 0.3633
• Accuracy: 0.8396
• Precision: 0.8128
• Recall: 0.7890
• F1: 0.7992

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
0.5664	1.0	122	0.5221	0.7218	0.6580	0.6432	0.6487
0.5148	2.0	244	0.5111	0.7243	0.6758	0.6899	0.6810
0.4924	3.0	366	0.4791	0.7444	0.6884	0.6741	0.6799
0.4615	4.0	488	0.4651	0.7644	0.7148	0.7058	0.7099
0.4516	5.0	610	0.4581	0.7644	0.7214	0.7408	0.7286
0.4291	6.0	732	0.4295	0.7895	0.7462	0.7385	0.7421
0.4194	7.0	854	0.4191	0.7995	0.7581	0.7606	0.7593
0.3994	8.0	976	0.4048	0.8120	0.7745	0.7645	0.7691
0.3919	9.0	1098	0.3950	0.8246	0.7954	0.7659	0.7778
0.3762	10.0	1220	0.3881	0.8271	0.8022	0.7626	0.7777
0.3704	11.0	1342	0.3806	0.8271	0.7949	0.7776	0.7853
0.3642	12.0	1464	0.3733	0.8421	0.8122	0.8008	0.8061
0.3614	13.0	1586	0.3753	0.8321	0.8092	0.7687	0.7842
0.3474	14.0	1708	0.3695	0.8396	0.8155	0.7840	0.7969
0.3479	15.0	1830	0.3675	0.8421	0.8142	0.7958	0.8040
0.3347	16.0	1952	0.3649	0.8421	0.8142	0.7958	0.8040
0.335	17.0	2074	0.3653	0.8371	0.8114	0.7822	0.7943
0.3361	18.0	2196	0.3632	0.8396	0.8128	0.7890	0.7992
0.3343	19.0	2318	0.3636	0.8371	0.8114	0.7822	0.7943
0.3347	20.0	2440	0.3633	0.8396	0.8128	0.7890	0.7992

Framework versions

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