README.md

metadata

library_name: transformers
license: mit
base_model: flax-community/indonesian-roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: indonesian-roberta-hate-speech-detection
    results: []

indonesian-roberta-hate-speech-detection

This model is a fine-tuned version of flax-community/indonesian-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4118
• Accuracy: 0.826
• Auc: 0.903
• Precision: 0.8257
• Recall: 0.8257
• F1: 0.8256

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc	Precision	Recall	F1
0.6653	1.0	188	0.5515	0.724	0.798	0.7243	0.7237	0.7238
0.5131	2.0	376	0.4578	0.783	0.875	0.7874	0.7830	0.7815
0.4307	3.0	564	0.4418	0.795	0.904	0.8137	0.7949	0.7928
0.3668	4.0	752	0.3804	0.835	0.916	0.8354	0.8349	0.8349
0.2991	5.0	940	0.4464	0.82	0.918	0.8345	0.8202	0.8190
0.2467	6.0	1128	0.4350	0.838	0.926	0.8482	0.8382	0.8375
0.2064	7.0	1316	0.4284	0.844	0.93	0.8515	0.8442	0.8438
0.1602	8.0	1504	0.4545	0.851	0.928	0.8520	0.8509	0.8509
0.1476	9.0	1692	0.4832	0.848	0.933	0.8545	0.8482	0.8479
0.117	10.0	1880	0.4632	0.855	0.932	0.8549	0.8549	0.8549
0.0925	11.0	2068	0.5501	0.86	0.933	0.8617	0.8602	0.8602
0.0811	12.0	2256	0.6097	0.849	0.933	0.8528	0.8489	0.8487
0.0702	13.0	2444	0.6568	0.85	0.933	0.8525	0.8495	0.8494
0.0588	14.0	2632	0.7092	0.844	0.93	0.8484	0.8442	0.8440
0.0572	15.0	2820	0.7012	0.855	0.933	0.8575	0.8549	0.8548
0.047	16.0	3008	0.7549	0.851	0.933	0.8545	0.8509	0.8507
0.0433	17.0	3196	0.7754	0.856	0.932	0.8593	0.8562	0.8561
0.044	18.0	3384	0.7868	0.85	0.932	0.8548	0.8502	0.8500
0.0408	19.0	3572	0.7890	0.855	0.932	0.8589	0.8549	0.8547
0.0412	20.0	3760	0.7870	0.855	0.932	0.8589	0.8549	0.8547

Framework versions

• Transformers 4.52.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1