tir-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: 13.7761
• F1-micro: 0.2768
• F1-macro: 0.2636

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: 36
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro
100.7132	1.0	92	13.9744	0.2324	0.2100
100.2803	2.0	184	13.8254	0.2798	0.2508
99.7058	3.0	276	13.7426	0.2907	0.2585
99.0804	4.0	368	13.7761	0.2768	0.2636

Framework versions

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