README.md

metadata

license: apache-2.0
base_model: bert-base-multilingual-uncased
tags:
  - generated_from_trainer
metrics:
  - recall
  - accuracy
model-index:
  - name: multibert_dataaugmentation
    results: []

multibert_dataaugmentation

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: 0.6354
• Precisions: 0.8725
• Recall: 0.8025
• F-measure: 0.8295
• Accuracy: 0.8996

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: 42
• 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
0.5999	1.0	284	0.4542	0.8507	0.7167	0.7422	0.8708
0.2808	2.0	568	0.3886	0.8552	0.7964	0.8192	0.8864
0.1596	3.0	852	0.4749	0.8786	0.7893	0.8208	0.8905
0.1035	4.0	1136	0.5060	0.8568	0.8034	0.8208	0.8954
0.0719	5.0	1420	0.5716	0.8498	0.8102	0.8255	0.8919
0.0456	6.0	1704	0.6255	0.8701	0.8060	0.8279	0.8960
0.0284	7.0	1988	0.6354	0.8725	0.8025	0.8295	0.8996
0.0205	8.0	2272	0.7146	0.8518	0.8105	0.8266	0.8988
0.011	9.0	2556	0.7307	0.8614	0.8045	0.8279	0.9004
0.0082	10.0	2840	0.7403	0.8785	0.7988	0.8255	0.9009
0.0064	11.0	3124	0.7756	0.8809	0.7913	0.8217	0.8989
0.0039	12.0	3408	0.8036	0.8650	0.7877	0.8130	0.8966
0.0038	13.0	3692	0.7660	0.8781	0.7950	0.8222	0.8997
0.0025	14.0	3976	0.7640	0.8829	0.7961	0.8249	0.9012

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1