Install the following python packages:
swifter
pandas
seaborn
nltk
sklearn

1. Run all the cells in the notebook
2. A pickel file (.pkl) will be created in the same directory 
3. The pickel file will be used in Creating API

Here, We only can have labels as 
0 - Geography/location
1 - Person
2 - date/time

Note: Inorder to identify more attributes frist we have to add the data in the dataset as provided and then again train the model and a new pickle file is generated.