sentiment-lora-r8

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: 0.2908
• Accuracy: 0.8772
• Precision: 0.8535
• Recall: 0.8481
• F1: 0.8507

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
0.556	1.0	122	0.5325	0.7168	0.6617	0.6671	0.6641
0.5103	2.0	244	0.4822	0.7719	0.7715	0.6386	0.6524
0.4637	3.0	366	0.4245	0.8045	0.7715	0.7342	0.7480
0.4173	4.0	488	0.3898	0.8246	0.7888	0.7859	0.7873
0.3674	5.0	610	0.3571	0.8371	0.8059	0.7947	0.7999
0.3484	6.0	732	0.3432	0.8371	0.8038	0.8022	0.8030
0.3247	7.0	854	0.3299	0.8521	0.8271	0.8079	0.8164
0.3102	8.0	976	0.3260	0.8622	0.8510	0.8050	0.8228
0.2991	9.0	1098	0.3138	0.8571	0.8349	0.8114	0.8216
0.29	10.0	1220	0.3123	0.8546	0.8324	0.8071	0.8180
0.2778	11.0	1342	0.3065	0.8672	0.8423	0.8335	0.8377
0.2702	12.0	1464	0.3006	0.8571	0.8349	0.8114	0.8216
0.2664	13.0	1586	0.2996	0.8596	0.8316	0.8282	0.8298
0.264	14.0	1708	0.2987	0.8722	0.8437	0.8521	0.8477
0.254	15.0	1830	0.2951	0.8772	0.8514	0.8531	0.8522
0.2571	16.0	1952	0.2945	0.8672	0.8463	0.8260	0.8351
0.2511	17.0	2074	0.2918	0.8722	0.8463	0.8446	0.8454
0.2574	18.0	2196	0.2909	0.8747	0.8510	0.8438	0.8473
0.2508	19.0	2318	0.2907	0.8772	0.8535	0.8481	0.8507
0.2536	20.0	2440	0.2908	0.8772	0.8535	0.8481	0.8507

Framework versions

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