fine_tuned_emBERT

This model is a fine-tuned version of morten-j/Mehdie_Extended-mBERT on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1441
• F1: 0.7568
• F5: 0.7196
• Precision: 0.875
• Recall: 0.6667

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 6
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	F5	Precision	Recall
No log	1.0	30	0.3540	0.0	0.0	0.0	0.0
No log	2.0	60	0.2975	0.4348	0.4077	0.5263	0.3704
No log	3.0	90	0.3550	0.45	0.3969	0.6923	0.3333
No log	4.0	120	0.2962	0.5405	0.4598	1.0	0.3704
No log	5.0	150	0.2629	0.5909	0.5437	0.7647	0.4815
No log	6.0	180	0.2459	0.5778	0.5368	0.7222	0.4815

Framework versions

• Transformers 4.38.1
• Pytorch 2.2.1+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2