sentiment-lora-r16-0

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: 0.3046
• Accuracy: 0.8672
• Precision: 0.8385
• Recall: 0.8435
• F1: 0.8409

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
0.5577	1.0	122	0.4974	0.7168	0.6466	0.6196	0.6267
0.4834	2.0	244	0.4640	0.7569	0.7260	0.7630	0.7327
0.4095	3.0	366	0.3775	0.8296	0.7937	0.7994	0.7964
0.339	4.0	488	0.3585	0.8446	0.8120	0.8151	0.8135
0.3189	5.0	610	0.3868	0.8296	0.7951	0.8294	0.8068
0.2953	6.0	732	0.3580	0.8496	0.8158	0.8436	0.8267
0.2737	7.0	854	0.3384	0.8571	0.8260	0.8339	0.8298
0.2691	8.0	976	0.3253	0.8647	0.8472	0.8167	0.8296
0.2496	9.0	1098	0.3504	0.8596	0.8278	0.8432	0.8347
0.2457	10.0	1220	0.3211	0.8596	0.8316	0.8282	0.8298
0.2386	11.0	1342	0.3201	0.8647	0.8387	0.8317	0.8351
0.2377	12.0	1464	0.3218	0.8672	0.8378	0.8460	0.8417
0.2277	13.0	1586	0.3138	0.8672	0.8393	0.8410	0.8402
0.2276	14.0	1708	0.3163	0.8647	0.8352	0.8417	0.8383
0.2271	15.0	1830	0.3158	0.8697	0.8399	0.8528	0.8458
0.2086	16.0	1952	0.3202	0.8647	0.8332	0.8517	0.8413
0.2151	17.0	2074	0.3024	0.8747	0.8510	0.8438	0.8473
0.2206	18.0	2196	0.3133	0.8672	0.8363	0.8535	0.8439
0.2044	19.0	2318	0.3063	0.8672	0.8378	0.8460	0.8417
0.2074	20.0	2440	0.3046	0.8672	0.8385	0.8435	0.8409

Framework versions

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