README.md

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-r8-2
    results: []

nerugm-lora-r8-2

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: 0.1743
• Precision: 0.6820
• Recall: 0.8289
• F1: 0.7483
• Accuracy: 0.9445

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
1.2665	1.0	106	0.7137	0.0	0.0	0.0	0.8449
0.713	2.0	212	0.6075	0.0	0.0	0.0	0.8451
0.6346	3.0	318	0.5231	0.1905	0.0118	0.0222	0.8494
0.5555	4.0	424	0.4458	0.275	0.0649	0.1050	0.8656
0.4696	5.0	530	0.3715	0.4802	0.2861	0.3586	0.8949
0.3932	6.0	636	0.3134	0.5563	0.5251	0.5402	0.9194
0.3299	7.0	742	0.2706	0.5968	0.6637	0.6285	0.9277
0.2896	8.0	848	0.2433	0.62	0.7316	0.6712	0.9340
0.2656	9.0	954	0.2277	0.6289	0.7699	0.6923	0.9355
0.2442	10.0	1060	0.2082	0.6526	0.7758	0.7089	0.9387
0.23	11.0	1166	0.2020	0.6390	0.7935	0.7079	0.9382
0.2229	12.0	1272	0.1977	0.6524	0.8083	0.7220	0.9385
0.2132	13.0	1378	0.1886	0.6602	0.8083	0.7268	0.9402
0.2055	14.0	1484	0.1810	0.6708	0.7994	0.7295	0.9415
0.2038	15.0	1590	0.1822	0.6595	0.8112	0.7275	0.9405
0.2004	16.0	1696	0.1788	0.6731	0.8201	0.7394	0.9430
0.1966	17.0	1802	0.1775	0.6731	0.8260	0.7417	0.9432
0.1931	18.0	1908	0.1765	0.6683	0.8260	0.7388	0.9435
0.1937	19.0	2014	0.1749	0.6747	0.8260	0.7427	0.9437
0.1888	20.0	2120	0.1743	0.6820	0.8289	0.7483	0.9445

Framework versions

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