2

I am using kafka-python to create a Kafka consumer (connected to Kafka producer, broker etc.) in Python that is receiving a live stream of data.

Now I just want to have a live timeseries plot of this data.

I found Plotly as very nice Python library but it seems like they are only providing some server-based solutions.

Can I make Plotly work for this offline? Can you suggest other libraries?

Thanks a lot!

BobBetter
  • 195
  • 3
  • 14
  • Plotly and matplotlib work offline just fine. I think you're confused that Dash, for example, starts its own Flask server... It'd be no different than a Jupyter notebook. What's the issue you're having when you try to use them? – OneCricketeer Oct 17 '18 at 12:17
  • Any of the examples for plotting a live stream of time series data were referencing the server-supported version ( https://plot.ly/python/streaming-tutorial/ ). And since I am new to Plotly I was not able to make the connection to "plotly.offline" . I am using Bokeh now, and that was much simpler to get started for me. Thanks anyways – BobBetter Oct 19 '18 at 14:25
  • Alright, basically the problem with Kafka (or any such message queue), is that you need to determine how/when you want to batch the records together such that you aren't overloading the downstream system with one record update at a time. – OneCricketeer Oct 19 '18 at 14:29
  • I used this tutorial: https://www.youtube.com/watch?v=NUrhOj3DzYs – BobBetter Oct 19 '18 at 17:18
  • @BobBetter the link to that YouTube video is dead. – andand Jul 27 '21 at 17:17

0 Answers0