bert-base-multilingual-uncased-finetuned-classification

This model is a fine-tuned version of google-bert/bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1394
• Accuracy: 0.9524

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	20	0.1029	0.9810
No log	2.0	40	0.1137	0.9524
No log	3.0	60	0.1153	0.9524
No log	4.0	80	0.1170	0.9524
No log	5.0	100	0.1208	0.9524
No log	6.0	120	0.1064	0.9810
No log	7.0	140	0.1344	0.9524
No log	8.0	160	0.1237	0.9524
No log	9.0	180	0.1146	0.9524
No log	10.0	200	0.1330	0.9524
No log	11.0	220	0.1285	0.9524
No log	12.0	240	0.1291	0.9524
No log	13.0	260	0.1335	0.9524
No log	14.0	280	0.1380	0.9524
No log	15.0	300	0.1394	0.9524

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2