Noctuaru/tone_berita_p2_v1

results

This model is a fine-tuned version of indobenchmark/indobert-base-p2 on the None dataset. It achieves the following results on the evaluation set:

• Accuracy: 0.8
• F1 Weighted: 0.7851
• Loss: 0.6697

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: 2.7820079535067715e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 17
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	F1 Weighted	Validation Loss
No log	1.0	6	0.68	0.6809	0.7999
0.6066	2.0	12	0.68	0.6398	0.7389
0.6066	3.0	18	0.76	0.7482	0.6278
0.5568	4.0	24	0.76	0.7332	0.7357
0.4146	5.0	30	0.8	0.7851	0.6697
0.4146	6.0	36	0.76	0.7510	0.6952
0.3241	7.0	42	0.76	0.7510	0.6229
0.3241	8.0	48	0.76	0.7510	0.7119
0.2655	9.0	54	0.76	0.7510	0.6794
0.195	10.0	60	0.76	0.7510	0.6919
0.195	11.0	66	0.76	0.7510	0.7807
0.1421	12.0	72	0.76	0.7510	0.7293
0.1421	13.0	78	0.76	0.7510	0.7324
0.1003	14.0	84	0.76	0.7510	0.8168
0.1141	15.0	90	0.76	0.7510	0.7625

Framework versions

• Transformers 4.56.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1