sentiment-lora-r8a1d0.05-1

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: 0.3148
• Accuracy: 0.8697
• Precision: 0.8474
• Recall: 0.8328
• F1: 0.8395

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
0.5657	1.0	122	0.5161	0.7243	0.6616	0.6474	0.6529
0.5088	2.0	244	0.4913	0.7393	0.6917	0.7056	0.6971
0.4682	3.0	366	0.4424	0.7845	0.7401	0.7425	0.7413
0.4114	4.0	488	0.3980	0.8095	0.7702	0.7702	0.7702
0.3862	5.0	610	0.3890	0.8145	0.7783	0.8088	0.7889
0.3512	6.0	732	0.3583	0.8496	0.8245	0.8036	0.8128
0.3428	7.0	854	0.3496	0.8521	0.8207	0.8254	0.8229
0.3254	8.0	976	0.3425	0.8496	0.8245	0.8036	0.8128
0.3226	9.0	1098	0.3388	0.8571	0.8310	0.8189	0.8245
0.3063	10.0	1220	0.3376	0.8647	0.8439	0.8217	0.8315
0.2939	11.0	1342	0.3319	0.8672	0.8463	0.8260	0.8351
0.2838	12.0	1464	0.3323	0.8546	0.8263	0.8196	0.8229
0.2916	13.0	1586	0.3283	0.8647	0.8472	0.8167	0.8296
0.2826	14.0	1708	0.3244	0.8672	0.8463	0.8260	0.8351
0.2739	15.0	1830	0.3231	0.8697	0.8449	0.8378	0.8412
0.2674	16.0	1952	0.3221	0.8697	0.8449	0.8378	0.8412
0.2648	17.0	2074	0.3193	0.8722	0.8528	0.8321	0.8413
0.2687	18.0	2196	0.3172	0.8697	0.8460	0.8353	0.8404
0.264	19.0	2318	0.3170	0.8747	0.8552	0.8363	0.8448
0.2637	20.0	2440	0.3148	0.8697	0.8474	0.8328	0.8395

Framework versions

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