How to automatically update the matplotlib graph and save it as PNG to a certain folder?

Question

I have a code that generates an overlaying/superimposed gantt chart with x-axis being time values and a vertical red line representing the current time.

In my current setup, my graph is generated from two dataframes (df and df1) and one of them (df1) is going to be constantly updated every hour or so (another dataframe stays static throughout the day and change only once at the end of it). I would like to save my plot in a folder and I want it to be automatically replaced by an updated graph in a scheduled intervals. To be clear, I want it to update/redraw the graph and replace the old PNG file in a folder with a newer graph.

I know there is a savefig function but it only saves my current plot once without replacing the old one. I googled around the web but couldn't find an appropriate solution. I am not even sure if that is possible. If it is not, I would really appreciate if you could advice an alternative solution for automatically updating the graph.

Here is the graph itself: Gantt Chart

The main goal of this is to show progress made through out the day and have more or less "live" graph. I already have an app deployed which takes the PNG file from a certain folder and displays it full screen and swaps "views" between two URL (2 web pages) and an assigned PNG folder.

Here is a part of my code that generates the chart. Let me know and I can provide sample data along with it for testing:

import pandas as pd
from datetime import datetime
import matplotlib.dates as mdates
import matplotlib.pyplot as plt
import math
import time
import matplotlib.patches as mpatches
df = pd.read_csv('zpp00141_new.csv')
df1 = pd.read_csv('sample_data1.csv')
#convert times to number to calculate
def timestr_to_num(timestr):
    return mdates.date2num(datetime.strptime('0' + timestr if timestr[1] == ':' else timestr, '%I:%M:%S %p'))
#gantt chart plotting
fig, ax = plt.subplots(figsize=(12.5, 6))
operations = pd.unique(df['Operation/Activity'])
#Declaring a colormap
cmap = plt.cm.get_cmap('plasma_r')
colors = [cmap(i/20) for i in range(20)]
#plot scheduled operations 
for operation, color in zip(operations, colors):
    for row in df[df['Operation/Activity'] == operation].itertuples():
        left = timestr_to_num(row.start)
        right = timestr_to_num(row.finish)
        ax.barh(operation, left=left, width=right - left, height=1, color="#6ED06F", edgecolor = 'black', label = operation)
#plot "Actual" operations' times     
operations1 = pd.unique(df1['Operation/Activity'])
for operation, color in zip(operations1, colors):
    for row in df1[df1['Operation/Activity'] == operation].itertuples():
        left = timestr_to_num(row.start)
        right = timestr_to_num(row.finish)
        ax.barh(operation, left=left, width=right - left, height=0.4, color='#33AFFF', edgecolor = 'black', label = operation)
        #plt.legend()
#set x-axis limit
ax.set_xlim(timestr_to_num('06:00:00 AM'), timestr_to_num('4:30:00 PM'))
ax.xaxis.set_major_formatter(mdates.DateFormatter('%H:%M'))  # display ticks as hours and minutes
ax.xaxis.set_major_locator(mdates.HourLocator(interval=1))  # set a tick every hour
#set y and x labels
ax.set_xlabel('Time', fontsize = 'large')
ax.set_ylabel('Operation', fontsize = 'large')
plt.tight_layout()
ax.set_yticklabels(list(df['Operation/Activity'] + " - " + df['Operation short text'].astype(str)))
ax.tick_params(axis='both', which='major', labelsize=11)
#add a title to a plot
plt.title("Schedule", fontsize = 'x-large')
#convert "now" time to a certain format of hours:minutes:seconds am/pm and plot a vertical red line for the "Current time"
now = datetime.now()
now = now.strftime('%I:%M:%S %p')
plt.axvline(x=timestr_to_num(now),color='r', animated = True)
#adding the legend
patch = mpatches.Patch(color='#33AFFF', label='Actual')
patch2 = mpatches.Patch(color='#6ED06F', label='Planned')
plt.legend(handles=[patch, patch2], loc='upper left', fontsize = 'large')
#plt.legend(handles=[patch2], loc='upper left', fontsize = 'x-large')
plt.show()

`fig.savefig('path/to/folder/chart.png')` will replace the old chart with the new one every time you run the script. How are You saving your graph? — jjsantoso, Mar 12 '20 at 15:08
@JuanJavierSantosOchoa I was using simple `fig.savefig('plot_test1.png, bbpx_inches='tight')`, I want my graph to be automatically saved without my interaction. I guess I do need to add another "loop-like" module to my code as Emerson suggested. But thanks a lot for a suggestion anyway! — Max, Mar 13 '20 at 12:35

score 0 · Answer 1 · answered Mar 12 '20 at 14:13

0

This script as you have shown will create a graph. What you need now is another layer in the module you want to create which will run on a regular interval, manage a file system, print the output of this plot to an image file, and then update the necessary files.

If the operating system you are running on is Linux, you could set up a linux Cron job to automatically run your image update function. You would also benefit from making a simple Bash script which could set up your virtual environment and then run the script.

Apart from those Linux functions, you will want to explore the basic tutorials for Pathlib. This built in python library will allow you to read in your file system, and make changes to the files.

Then to plot the PNG files, reference this existing Stack question

answered Mar 12 '20 at 14:13

Emerson Maki

44
1
5

thanks for a response, unfortunately, I am using Windows 10 and I cannot set up Linux due to company's IT policies. I will definitely check out the Pathlib. As for the other layer in the module, maybe you could provide a brief description of what should I implement in the module or write a pseudo code by any chance? I am quite new to Python and would really appreciate any assistance. – Max Mar 13 '20 at 12:40
I understand the IT restrictions. Cron is just a type of scheduled command for which windows 10 should have an equivalent program. [Here is an existing discussion on the matter](https://stackoverflow.com/questions/7195503/setting-up-a-cron-job-in-windows) How I imagine you set up this script goes something like this: - Check and set up any missing directories - Run your .png file generator in temp directory - Once successful generation of .PNG, Move previous .PNG file into archive file with date in name - Move the brand new .PNG file into primary directory - Program Complete. – Emerson Maki Mar 13 '20 at 17:27

How to automatically update the matplotlib graph and save it as PNG to a certain folder?

1 Answers1