indonesia-election-topic-classification-undersampling-double-largee

This model is a fine-tuned version of indobenchmark/indobert-large-p2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0803
• Balanced Accuracy: 0.5882

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Balanced Accuracy
No log	1.0	180	1.1464	0.4620
No log	2.0	360	1.0442	0.5413
1.0029	3.0	540	1.3242	0.5189
1.0029	4.0	720	1.5257	0.5678
1.0029	5.0	900	1.8255	0.5412
0.2989	6.0	1080	2.0509	0.5727
0.2989	7.0	1260	2.0685	0.5875
0.2989	8.0	1440	2.1902	0.5759
0.1005	9.0	1620	2.1111	0.5553
0.1005	10.0	1800	2.0803	0.5882
0.1005	11.0	1980	2.1430	0.5653
0.0526	12.0	2160	2.1122	0.5743
0.0526	13.0	2340	2.1340	0.5737
0.0342	14.0	2520	2.1476	0.5737

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1