sentiment-unipelt-4

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: 0.2986
• Accuracy: 0.8772
• Precision: 0.8561
• Recall: 0.8431
• F1: 0.8492

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
0.552	1.0	122	0.5144	0.7293	0.6759	0.6810	0.6782
0.4663	2.0	244	0.4076	0.8170	0.8013	0.7330	0.7540
0.37	3.0	366	0.3749	0.8346	0.8484	0.7405	0.7687
0.3284	4.0	488	0.3257	0.8571	0.8273	0.8289	0.8281
0.3035	5.0	610	0.3099	0.8546	0.8246	0.8246	0.8246
0.2705	6.0	732	0.2940	0.8697	0.8474	0.8328	0.8395
0.2439	7.0	854	0.2855	0.8697	0.8488	0.8303	0.8386
0.2232	8.0	976	0.3066	0.8722	0.8528	0.8321	0.8413
0.2142	9.0	1098	0.2912	0.8747	0.8625	0.8263	0.8413
0.2009	10.0	1220	0.2746	0.8546	0.8254	0.8221	0.8238
0.1825	11.0	1342	0.2736	0.8647	0.8359	0.8392	0.8376
0.1717	12.0	1464	0.2938	0.8622	0.8399	0.8200	0.8289
0.1634	13.0	1586	0.2855	0.8747	0.8457	0.8588	0.8517
0.158	14.0	1708	0.2812	0.8772	0.8514	0.8531	0.8522
0.1528	15.0	1830	0.2863	0.8747	0.8479	0.8513	0.8496
0.1486	16.0	1952	0.2951	0.8722	0.8512	0.8346	0.8422
0.1455	17.0	2074	0.2931	0.8822	0.8596	0.8542	0.8568
0.1318	18.0	2196	0.2960	0.8772	0.8535	0.8481	0.8507
0.1287	19.0	2318	0.2975	0.8772	0.8561	0.8431	0.8492
0.1243	20.0	2440	0.2986	0.8772	0.8561	0.8431	0.8492

Framework versions

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