sun-mbert-focal

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

• Loss: 21.6982
• F1-micro: 0.4824
• F1-macro: 0.3694

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 9
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro
157.0262	1.0	24	22.4103	0.3459	0.3039
147.2237	2.0	48	21.9937	0.3519	0.2944
146.3563	3.0	72	21.7622	0.4937	0.3721
142.832	4.0	96	21.6982	0.4824	0.3694

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1