metadata

language:
  - id
license: mit
base_model: indolem/indobert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: nerugm-lora-r8a0d0.1
    results: []

nerugm-lora-r8a0d0.1

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: 0.1278
Precision: 0.7600
Recall: 0.8815
F1: 0.8162
Accuracy: 0.9593

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
0.713	1.0	528	0.3558	0.4950	0.3736	0.4258	0.8990
0.2793	2.0	1056	0.1931	0.6472	0.8048	0.7174	0.9392
0.1876	3.0	1584	0.1619	0.6758	0.8466	0.7516	0.9462
0.1593	4.0	2112	0.1416	0.7366	0.8629	0.7948	0.9555
0.1412	5.0	2640	0.1350	0.7386	0.8652	0.7969	0.9559
0.1325	6.0	3168	0.1361	0.7324	0.8698	0.7952	0.9555
0.126	7.0	3696	0.1383	0.7310	0.8698	0.7944	0.9553
0.1194	8.0	4224	0.1349	0.7456	0.8838	0.8088	0.9583
0.1137	9.0	4752	0.1299	0.7495	0.8745	0.8072	0.9583
0.1112	10.0	5280	0.1285	0.7455	0.8698	0.8029	0.9579
0.1065	11.0	5808	0.1304	0.7525	0.8815	0.8119	0.9587
0.1044	12.0	6336	0.1329	0.7520	0.8791	0.8106	0.9577
0.1026	13.0	6864	0.1257	0.7520	0.8722	0.8076	0.9585
0.0989	14.0	7392	0.1265	0.7626	0.8791	0.8167	0.9599
0.0982	15.0	7920	0.1281	0.7631	0.8815	0.8180	0.9597
0.0974	16.0	8448	0.1264	0.7515	0.8768	0.8093	0.9597
0.0966	17.0	8976	0.1282	0.7545	0.8838	0.8140	0.9589
0.095	18.0	9504	0.1292	0.7570	0.8815	0.8145	0.9589
0.0941	19.0	10032	0.1268	0.7585	0.8815	0.8154	0.9595
0.0948	20.0	10560	0.1278	0.7600	0.8815	0.8162	0.9593

Framework versions

Transformers 4.39.3
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.15.2