The document contains the code piece which converts an input pandas dataframe into a kafka stream. This code can be used with any custom block created in the RZT AI Platform.

Block Code

# Creating an Output Topic and a Producer Object. 
# This should be written inside run method.

output_topic,producer=self.data_handler.create_producer(display_name="< TopicAlias Name>")

# The following lines of code should be written inside stream method.
# A List containing the names of all columns of the dataframe. For e.g. columns = ["C1", "C2", "C3", "C4", "C5"]

columns=[< ColumnNames>]

# Iterating over the rows in Pandas Dataframe and sending it to Output Topic 
# Here df is the Pandas Dataframe

for index, row in df.iterrows(): producer.send([row[x] for x in columns])

# Closing the Producer Object