MBERTdataaugmentation1511

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

• Loss: 0.4799
• Precisions: 0.8775
• Recall: 0.8066
• F-measure: 0.8358
• Accuracy: 0.9364

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: 7.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 34
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Precisions	Recall	F-measure	Accuracy
0.4277	1.0	284	0.2782	0.7721	0.7565	0.7579	0.9183
0.1776	2.0	568	0.2903	0.8401	0.7736	0.8007	0.9265
0.0953	3.0	852	0.3116	0.8565	0.7792	0.8124	0.9337
0.067	4.0	1136	0.3747	0.8019	0.7974	0.7931	0.9253
0.0424	5.0	1420	0.3834	0.8297	0.7751	0.7971	0.9306
0.0288	6.0	1704	0.4208	0.8641	0.7981	0.8245	0.9313
0.0187	7.0	1988	0.4472	0.8374	0.8033	0.8148	0.9323
0.0105	8.0	2272	0.4297	0.8652	0.7903	0.8181	0.9332
0.0085	9.0	2556	0.4425	0.8567	0.8019	0.8250	0.9344
0.0044	10.0	2840	0.4744	0.8493	0.8075	0.8256	0.9347
0.004	11.0	3124	0.4967	0.8581	0.8017	0.8251	0.9361
0.003	12.0	3408	0.4799	0.8775	0.8066	0.8358	0.9364
0.0015	13.0	3692	0.5093	0.8803	0.7943	0.8294	0.9356
0.0014	14.0	3976	0.4985	0.8762	0.7964	0.8281	0.9359

Framework versions

• Transformers 4.35.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.14.1