A pre-trained GRU model classifies the Romanized Chinese names into [1] names from Hong Kong ; [2] names from Mainland China; and [3] Company's names.
The raw data is not available in Github because of privacy concerns. Users can checkout 01_proc_data.ipynb
and 02_Model training.ipynb
for example data.
02_Model training.ipynb
train the GRU model with processed data. The pre-trained model is availability in the model
folder.
Please check out 03_Example.ipynb
for usage of the pre-trained model.