After creating the data structure of tabular format mentioned above we will be calculating the predicted answer to a question every time a user queries our bot. We load all the question answers pair from the dataset. Let's load our CSV file using pandas and perform some pre-processing on the dataset:
Code Link: https://github.com/PacktPublishing/Python-Deep-Learning-Projects/tree/master/Chapter4/tfidf_version
import pandas as pd
filepath = 'sample_data.csv'
csv_reader=pd.read_csv(filepath)
question_list = csv_reader[csv_reader.columns[0]].values.tolist()
answers_list = csv_reader[csv_reader.columns[1]].values.tolist()
query= 'Can I get an Americano, btw how much it will cost ?'