metadata

library_name: transformers
license: mit
base_model: w11wo/indonesian-roberta-base-sentiment-classifier
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: results_final
    results: []

results_final

This model is a fine-tuned version of w11wo/indonesian-roberta-base-sentiment-classifier on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.0710
Accuracy: 0.9073
F1 Macro: 0.9073
F1 Weighted: 0.9073
Precision Macro: 0.9074
Recall Macro: 0.9072
Precision Weighted: 0.9074
Recall Weighted: 0.9073

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: 8.550119000665763e-05
train_batch_size: 8
eval_batch_size: 16
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: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 13

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	F1 Weighted	Precision Macro	Recall Macro	Precision Weighted	Recall Weighted
0.0	1.8182	500	1.0710	0.9073	0.9073	0.9073	0.9074	0.9072	0.9074	0.9073
0.0281	3.6364	1000	1.0771	0.8927	0.8930	0.8931	0.8945	0.8927	0.8946	0.8927
0.0103	5.4545	1500	0.9934	0.8909	0.8910	0.8910	0.8914	0.8909	0.8915	0.8909
0.0763	7.2727	2000	0.8136	0.8891	0.8889	0.8890	0.8889	0.8890	0.8890	0.8891
0.0468	9.0909	2500	0.8967	0.8836	0.8835	0.8835	0.8855	0.8834	0.8854	0.8836
0.0429	10.9091	3000	0.9118	0.8909	0.8909	0.8909	0.8914	0.8908	0.8914	0.8909
0.0374	12.7273	3500	0.9128	0.8964	0.8962	0.8963	0.8965	0.8962	0.8965	0.8964

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 2.14.4
Tokenizers 0.21.1