fine-tuned-DatasetQAS-Squad-ID-with-indobert-base-uncased-with-ITTL-without-freeze-LR-1e-05

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

• Loss: 1.5061
• Exact Match: 48.9695
• F1: 65.3139

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-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 64
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Exact Match	F1
2.0131	0.5	463	1.8398	39.7325	55.0682
1.8343	1.0	926	1.6799	42.9545	58.8261
1.6407	1.5	1389	1.6097	45.1502	61.1235
1.6021	2.0	1852	1.5634	46.0167	62.5172
1.4841	2.5	2315	1.5438	46.5971	63.4037
1.5117	3.0	2778	1.5080	47.2028	64.1405
1.3752	3.5	3241	1.5149	47.6487	64.3195
1.3491	4.0	3704	1.4956	47.8927	64.4993
1.3068	4.5	4167	1.4951	48.0861	64.7876
1.2916	5.0	4630	1.4895	48.3722	64.9676
1.2593	5.5	5093	1.4954	48.5909	65.1206
1.2032	6.0	5556	1.4891	48.5236	65.0831
1.1826	6.5	6019	1.4944	48.6077	65.0162
1.2159	7.0	6482	1.4870	48.9526	65.1941
1.1503	7.5	6945	1.5074	48.8180	65.3672
1.1683	8.0	7408	1.4928	48.7760	65.2063
1.0898	8.5	7871	1.5141	48.7844	65.0996
1.1217	9.0	8334	1.5061	48.9695	65.3139

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu117
• Datasets 2.2.0
• Tokenizers 0.13.2