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-r8-4
    results: []

nerugm-lora-r8-4

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: 0.1784
Precision: 0.6597
Recall: 0.8120
F1: 0.7280
Accuracy: 0.9434

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
1.2537	1.0	106	0.7370	0.0	0.0	0.0	0.8366
0.7093	2.0	212	0.6298	0.1667	0.0028	0.0056	0.8373
0.6232	3.0	318	0.5443	0.2727	0.0171	0.0322	0.8417
0.5363	4.0	424	0.4559	0.3301	0.0969	0.1498	0.8624
0.4591	5.0	530	0.3863	0.4744	0.2906	0.3604	0.8929
0.387	6.0	636	0.3272	0.5789	0.5641	0.5714	0.9190
0.3252	7.0	742	0.2811	0.6035	0.6895	0.6436	0.9291
0.2874	8.0	848	0.2455	0.6108	0.7066	0.6552	0.9313
0.2588	9.0	954	0.2285	0.6179	0.7464	0.6761	0.9333
0.2393	10.0	1060	0.2153	0.6362	0.7721	0.6976	0.9363
0.224	11.0	1166	0.2062	0.6339	0.7892	0.7030	0.9387
0.2137	12.0	1272	0.2002	0.6473	0.7949	0.7136	0.9387
0.2052	13.0	1378	0.1889	0.6611	0.7949	0.7219	0.9424
0.2039	14.0	1484	0.1862	0.6690	0.8063	0.7313	0.9431
0.1975	15.0	1590	0.1868	0.6682	0.8091	0.7320	0.9431
0.1936	16.0	1696	0.1837	0.6628	0.8177	0.7321	0.9427
0.1908	17.0	1802	0.1825	0.6620	0.8148	0.7305	0.9427
0.1885	18.0	1908	0.1806	0.6582	0.8120	0.7270	0.9431
0.1877	19.0	2014	0.1783	0.6581	0.8063	0.7247	0.9431
0.1858	20.0	2120	0.1784	0.6597	0.8120	0.7280	0.9434

Framework versions

Transformers 4.39.3
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.15.2