roberta-base-indonesian-522M-with-sa-william-dataset

This model is a fine-tuned version of cahya/roberta-base-indonesian-522M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0934

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
0.2392	1.0	65	0.1149
0.1242	2.0	130	0.1343
0.0749	3.0	195	0.0934
0.0522	4.0	260	0.2300
0.0333	5.0	325	0.2182
0.027	6.0	390	0.1472
0.0115	7.0	455	0.1757
0.0162	8.0	520	0.1539
0.0219	9.0	585	0.1905
0.0084	10.0	650	0.1709
0.0014	11.0	715	0.2220
0.0004	12.0	780	0.2575
0.0002	13.0	845	0.2390
0.0	14.0	910	0.2316
0.0	15.0	975	0.2333
0.0	16.0	1040	0.2346
0.0	17.0	1105	0.2357
0.0	18.0	1170	0.2366
0.0	19.0	1235	0.2371
0.0	20.0	1300	0.2372

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2