Zidan_model_output_v1

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

• Loss: 0.5962
• Accuracy: 0.7636

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-06
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	248	1.0058	0.5818
No log	2.0	496	0.9174	0.6182
0.9971	3.0	744	0.8306	0.6182
0.9971	4.0	992	0.7732	0.6727
0.8391	5.0	1240	0.7360	0.6545
0.8391	6.0	1488	0.7084	0.6909
0.7312	7.0	1736	0.6780	0.6909
0.7312	8.0	1984	0.6657	0.7455
0.6526	9.0	2232	0.6486	0.7455
0.6526	10.0	2480	0.6153	0.7818
0.5949	11.0	2728	0.6290	0.7818
0.5949	12.0	2976	0.6071	0.7455
0.5579	13.0	3224	0.6088	0.7818
0.5579	14.0	3472	0.5985	0.7636
0.5308	15.0	3720	0.5962	0.7636

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1