rus-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: 7.6542
• F1-micro: 0.5565
• F1-macro: 0.5564

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: 26
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro
95.0121	1.0	67	10.0905	0.4495	0.4478
76.5704	2.0	134	8.0194	0.5158	0.5161
67.4816	3.0	201	7.7426	0.5409	0.5376
63.3027	4.0	268	7.6542	0.5565	0.5564

Framework versions

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