nerugm-lora-r4a2d0.15

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: 0.1301
• Precision: 0.7357
• Recall: 0.8652
• F1: 0.7952
• Accuracy: 0.9577

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: 16
• eval_batch_size: 64
• 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	Precision	Recall	F1	Accuracy
0.7663	1.0	528	0.4380	0.3934	0.1116	0.1738	0.8659
0.3481	2.0	1056	0.2220	0.6018	0.7403	0.6639	0.9339
0.2139	3.0	1584	0.1790	0.6561	0.8327	0.7339	0.9400
0.1777	4.0	2112	0.1535	0.7164	0.8559	0.7800	0.9512
0.1578	5.0	2640	0.1445	0.7367	0.8698	0.7978	0.9535
0.1469	6.0	3168	0.1441	0.7139	0.8745	0.7861	0.9535
0.1399	7.0	3696	0.1453	0.7175	0.8838	0.7920	0.9524
0.1333	8.0	4224	0.1403	0.7298	0.8838	0.7995	0.9547
0.1273	9.0	4752	0.1368	0.7387	0.8722	0.7999	0.9563
0.1246	10.0	5280	0.1342	0.7426	0.8768	0.8042	0.9569
0.1195	11.0	5808	0.1351	0.7359	0.8791	0.8012	0.9571
0.1172	12.0	6336	0.1349	0.7373	0.8791	0.8020	0.9573
0.1155	13.0	6864	0.1296	0.7441	0.8768	0.8050	0.9581
0.1118	14.0	7392	0.1302	0.7367	0.8698	0.7978	0.9577
0.1111	15.0	7920	0.1322	0.7426	0.8768	0.8042	0.9577
0.1097	16.0	8448	0.1303	0.7353	0.8698	0.7969	0.9577
0.1094	17.0	8976	0.1306	0.7343	0.8722	0.7973	0.9573
0.1077	18.0	9504	0.1319	0.7372	0.8722	0.7990	0.9577
0.1065	19.0	10032	0.1296	0.7376	0.8675	0.7973	0.9577
0.1078	20.0	10560	0.1301	0.7357	0.8652	0.7952	0.9577

Framework versions

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