yemen2016/nbbert_ED2

nbbert_ED1

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

• Loss: 0.4147
• F1-score: 0.8769

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1-score
No log	1.0	69	0.7063	0.3425
No log	2.0	138	0.6562	0.4700
No log	3.0	207	0.5758	0.8114
No log	4.0	276	0.4802	0.8441
No log	5.0	345	0.4557	0.8096
No log	6.0	414	0.4620	0.8597
No log	7.0	483	0.4147	0.8769
0.4906	8.0	552	0.5979	0.8442
0.4906	9.0	621	0.6290	0.8432
0.4906	10.0	690	0.5401	0.8443
0.4906	11.0	759	0.5805	0.8606
0.4906	12.0	828	0.6075	0.8688
0.4906	13.0	897	0.7802	0.8436
0.4906	14.0	966	0.7530	0.8432
0.1795	15.0	1035	0.6979	0.8606
0.1795	16.0	1104	0.7619	0.8524
0.1795	17.0	1173	0.7760	0.8525
0.1795	18.0	1242	0.8060	0.8525
0.1795	19.0	1311	0.8363	0.8525
0.1795	20.0	1380	0.8305	0.8525

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1