bert-base-multilingual-cased_0319_J

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

• Loss: 0.0355
• Precision: 0.9776
• Recall: 0.9791
• F1: 0.9784
• Accuracy: 0.9949

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: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	95	0.2410	0.6320	0.7491	0.6856	0.9475
No log	2.0	190	0.0483	0.9430	0.9522	0.9476	0.9912
No log	3.0	285	0.0379	0.9710	0.9746	0.9728	0.9938
No log	4.0	380	0.0382	0.9645	0.9731	0.9688	0.9940
No log	5.0	475	0.0357	0.9703	0.9761	0.9732	0.9941
No log	6.0	570	0.0367	0.9710	0.9761	0.9736	0.9943
No log	7.0	665	0.0376	0.9732	0.9761	0.9746	0.9943
No log	8.0	760	0.0355	0.9776	0.9791	0.9784	0.9949
No log	9.0	855	0.0364	0.9718	0.9768	0.9743	0.9946
No log	10.0	950	0.0361	0.9747	0.9776	0.9761	0.9947

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.0+cu117
• Datasets 2.8.0
• Tokenizers 0.12.1