sentiment-lora-r4a1d0.15-1

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: 0.3312
• Accuracy: 0.8622
• Precision: 0.8414
• Recall: 0.8175
• F1: 0.8279

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
0.566	1.0	122	0.5206	0.7143	0.6484	0.6353	0.6403
0.5117	2.0	244	0.5062	0.7343	0.6880	0.7045	0.6939
0.4804	3.0	366	0.4667	0.7669	0.7182	0.7126	0.7152
0.4345	4.0	488	0.4350	0.7920	0.7494	0.7403	0.7445
0.4081	5.0	610	0.4337	0.7945	0.7565	0.7846	0.7661
0.3793	6.0	732	0.3923	0.8195	0.7857	0.7673	0.7753
0.3665	7.0	854	0.3765	0.8296	0.7949	0.7919	0.7934
0.3471	8.0	976	0.3681	0.8371	0.8089	0.7872	0.7966
0.3498	9.0	1098	0.3677	0.8321	0.8024	0.7812	0.7904
0.3282	10.0	1220	0.3634	0.8346	0.8074	0.7805	0.7917
0.3149	11.0	1342	0.3537	0.8446	0.8180	0.7976	0.8065
0.3092	12.0	1464	0.3529	0.8496	0.8202	0.8136	0.8167
0.3135	13.0	1586	0.3471	0.8521	0.8332	0.7979	0.8122
0.3103	14.0	1708	0.3427	0.8622	0.8430	0.8150	0.8269
0.2974	15.0	1830	0.3372	0.8622	0.8385	0.8225	0.8298
0.2905	16.0	1952	0.3345	0.8697	0.8488	0.8303	0.8386
0.2895	17.0	2074	0.3339	0.8622	0.8430	0.8150	0.8269
0.2922	18.0	2196	0.3319	0.8697	0.8488	0.8303	0.8386
0.2843	19.0	2318	0.3319	0.8622	0.8430	0.8150	0.8269
0.287	20.0	2440	0.3312	0.8622	0.8414	0.8175	0.8279

Framework versions

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