sentiment-lora-r8-3

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: 0.2951
• Accuracy: 0.8722
• Precision: 0.8512
• Recall: 0.8346
• F1: 0.8422

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
0.5643	1.0	122	0.5218	0.7043	0.6241	0.5932	0.5982
0.5086	2.0	244	0.5021	0.7293	0.6814	0.6960	0.6868
0.4652	3.0	366	0.4450	0.7895	0.7464	0.7360	0.7407
0.4248	4.0	488	0.3932	0.8346	0.8074	0.7805	0.7917
0.3812	5.0	610	0.3704	0.8421	0.8083	0.8158	0.8119
0.3506	6.0	732	0.3566	0.8571	0.8266	0.8314	0.8289
0.3323	7.0	854	0.3438	0.8571	0.8365	0.8089	0.8206
0.3108	8.0	976	0.3326	0.8622	0.8414	0.8175	0.8279
0.2998	9.0	1098	0.3250	0.8672	0.8412	0.8360	0.8385
0.2923	10.0	1220	0.3182	0.8571	0.8289	0.8239	0.8264
0.2887	11.0	1342	0.3145	0.8722	0.8485	0.8396	0.8438
0.2716	12.0	1464	0.3092	0.8722	0.8498	0.8371	0.8430
0.2598	13.0	1586	0.3099	0.8772	0.8628	0.8331	0.8458
0.2722	14.0	1708	0.3003	0.8772	0.8561	0.8431	0.8492
0.2536	15.0	1830	0.2978	0.8772	0.8561	0.8431	0.8492
0.2536	16.0	1952	0.2970	0.8822	0.8596	0.8542	0.8568
0.2479	17.0	2074	0.2978	0.8822	0.8639	0.8467	0.8545
0.2487	18.0	2196	0.2970	0.8772	0.8561	0.8431	0.8492
0.2457	19.0	2318	0.2947	0.8722	0.8512	0.8346	0.8422
0.2499	20.0	2440	0.2951	0.8722	0.8512	0.8346	0.8422

Framework versions

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