sentiment-lora-r8a2d0.05-0

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: 0.3260
• Accuracy: 0.8622
• Precision: 0.8319
• Recall: 0.8400
• F1: 0.8357

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
0.5609	1.0	122	0.5086	0.7193	0.6580	0.6514	0.6543
0.4986	2.0	244	0.4855	0.7494	0.7127	0.7427	0.7201
0.4593	3.0	366	0.4238	0.7694	0.7249	0.7394	0.7309
0.3957	4.0	488	0.3916	0.8070	0.7670	0.7735	0.7700
0.3658	5.0	610	0.4266	0.7995	0.7641	0.7981	0.7744
0.3345	6.0	732	0.3666	0.8371	0.8028	0.8072	0.8049
0.3237	7.0	854	0.3714	0.8396	0.8045	0.8265	0.8136
0.304	8.0	976	0.3537	0.8421	0.8083	0.8158	0.8119
0.3027	9.0	1098	0.3531	0.8446	0.8111	0.8201	0.8153
0.2962	10.0	1220	0.3382	0.8521	0.8220	0.8204	0.8212
0.2721	11.0	1342	0.3490	0.8496	0.8162	0.8311	0.8229
0.2693	12.0	1464	0.3502	0.8546	0.8220	0.8372	0.8288
0.2745	13.0	1586	0.3284	0.8571	0.8289	0.8239	0.8264
0.2712	14.0	1708	0.3297	0.8596	0.8299	0.8332	0.8315
0.256	15.0	1830	0.3357	0.8647	0.8346	0.8442	0.8391
0.2504	16.0	1952	0.3346	0.8571	0.8255	0.8364	0.8306
0.2487	17.0	2074	0.3242	0.8571	0.8281	0.8264	0.8272
0.2514	18.0	2196	0.3309	0.8622	0.8314	0.8425	0.8365
0.2451	19.0	2318	0.3243	0.8622	0.8333	0.8350	0.8341
0.2461	20.0	2440	0.3260	0.8622	0.8319	0.8400	0.8357

Framework versions

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