I have a log file contains numbers indexed in time, I want to generate time series chart to display the data to my users. However I’m new to data visualization, so I’m tracking what I did in this post.

Environment Setup

Jupyter is a nice tool to playaround with python, so get it installed or get a docker instance:

docker run --rm -p 8888:8888 -v "$PWD":/Users/gouliang/jupyter-home jupyter/scipy-notebook

if you’re not sure about which image to choose: https://jupyter-docker-stacks.readthedocs.io/en/latest/using/selecting.html

url with access token will be displayed on the console.

Hello world

start a python notebook and run:

## https://stackoverflow.com/questions/19079143/how-to-plot-time-series-in-python

import matplotlib.pyplot as plt
import datetime
import numpy as np

x = np.array([datetime.datetime(2013, 9, 28, i, 0) for i in range(24)])
y = np.random.randint(100, size=x.shape)


a nice time series chart will be displayed.