NLP_91_1

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: 0.4408
• Accuracy: 0.9220
• Precision: 0.9156
• Recall: 0.9170
• F1: 0.9158

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.309	1.0	48	0.4280	0.8532	0.8506	0.8461	0.8454
0.2595	2.0	96	0.4335	0.8807	0.8767	0.8766	0.8738
0.2196	3.0	144	0.3883	0.8945	0.8956	0.8869	0.8876
0.1812	4.0	192	0.4664	0.8761	0.8856	0.8614	0.8638
0.1256	5.0	240	0.4764	0.8670	0.8750	0.8627	0.8625
0.142	6.0	288	0.5258	0.8670	0.8818	0.8580	0.8607
0.1006	7.0	336	0.4323	0.9037	0.8961	0.8989	0.8970
0.0897	8.0	384	0.4659	0.8991	0.8959	0.8891	0.8914
0.0595	9.0	432	0.4569	0.9174	0.9149	0.9099	0.9115
0.0399	10.0	480	0.4592	0.9037	0.8981	0.8970	0.8966
0.056	11.0	528	0.4461	0.9174	0.9102	0.9091	0.9094
0.0451	12.0	576	0.4772	0.8991	0.8926	0.8891	0.8906
0.0309	13.0	624	0.4396	0.9220	0.9160	0.9169	0.9155
0.0338	14.0	672	0.4423	0.9220	0.9156	0.9170	0.9158
0.0458	15.0	720	0.4408	0.9220	0.9156	0.9170	0.9158

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1