sentiment-lora-r8a2d0.15-1

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: 0.3168
• Accuracy: 0.8722
• Precision: 0.8528
• Recall: 0.8321
• F1: 0.8413

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
0.5647	1.0	122	0.5166	0.7068	0.6380	0.6250	0.6297
0.5067	2.0	244	0.4954	0.7343	0.6870	0.7020	0.6926
0.4617	3.0	366	0.4391	0.7920	0.7491	0.7503	0.7497
0.4044	4.0	488	0.3911	0.8145	0.7761	0.7788	0.7774
0.382	5.0	610	0.3827	0.8195	0.7849	0.8198	0.7962
0.3494	6.0	732	0.3528	0.8421	0.8092	0.8108	0.8100
0.3423	7.0	854	0.3442	0.8546	0.8239	0.8272	0.8255
0.33	8.0	976	0.3400	0.8672	0.8479	0.8235	0.8342
0.3296	9.0	1098	0.3349	0.8496	0.8245	0.8036	0.8128
0.3074	10.0	1220	0.3349	0.8622	0.8467	0.8100	0.8249
0.2911	11.0	1342	0.3240	0.8697	0.8503	0.8278	0.8377
0.2855	12.0	1464	0.3273	0.8722	0.8463	0.8446	0.8454
0.2903	13.0	1586	0.3285	0.8647	0.8472	0.8167	0.8296
0.2896	14.0	1708	0.3254	0.8672	0.8479	0.8235	0.8342
0.2744	15.0	1830	0.3241	0.8647	0.8377	0.8342	0.8359
0.2691	16.0	1952	0.3210	0.8571	0.8289	0.8239	0.8264
0.2671	17.0	2074	0.3208	0.8697	0.8503	0.8278	0.8377
0.2736	18.0	2196	0.3179	0.8722	0.8512	0.8346	0.8422
0.2662	19.0	2318	0.3180	0.8722	0.8544	0.8296	0.8404
0.2664	20.0	2440	0.3168	0.8722	0.8528	0.8321	0.8413

Framework versions

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