QUESTION: How to hvplot from csv file updated every 1 second

Hello I have a very simple use case. I have the following python script that just generates CPU usage of my machine every second and saves the data into a csv file. I am trying to use the streamz library to then use this csv file as a source data and plot the data using hvplot library.

Here’s my script for generating CPU usage date:

# pip install psutil
import csv
import psutil
import time

x_value = 0

fieldnames = ["x_value", "cpu_perc"]


with open('data.csv', 'w') as csv_file:
    csv_writer = csv.DictWriter(csv_file, fieldnames=fieldnames)
    csv_writer.writeheader()

while True:

    with open('data.csv', 'a') as csv_file:
        csv_writer = csv.DictWriter(csv_file, fieldnames=fieldnames)
        y_value = psutil.cpu_percent()

        info = {
            "x_value": x_value,
            "cpu_perc": y_value
        }

        csv_writer.writerow(info)
        print(x_value, y_value)

        x_value += 1

    time.sleep(1)

Then in my jupyter notebook, I’ve tried:

import hvplot.streamz
import pandas as pd
from streamz import Stream

source = Stream.from_textfile('data/data.csv', poll_interval=1)
source.start()
example = pd.DataFrame({'x': [], 'cpu_perc': []})
sdf = source.to_dataframe(example=example)
sdf.hvplot()

But I get an emtpy bokeh chart. I looked at the dataframe page in the documentation, but doesn’t really show how to use a csv file. I looked at the API docs and found the from_textfile() method from the Stream class and also found the to_dataframe() method, and just went from there.

[martindurant]

@chinmaychandak - an excellent example of the documentation being inadequate?

Perhaps it would be nice to have some sort of debug mode where, as an event travels through the graph, you can see what becomes of it at each node.

[jbednar]

Sort of, though my version (tornado_data_stream.ipynb in https://anaconda.org/jbednar/project/streaming_hvplot) doesn’t use a CSV file. But I think we were about to make a new release of streamz that would make it a whole lot simpler.