how to use FuncAnimation to update and animate multiple figures with matplotlib?

Question

Trying to create a program that reads serial data and updates multiple figures (1 line and 2 bar charts for now but could potentially be more).

Using 3 separate calls to FuncAnimation() right now, but proving to be really slow which is not good as I still need the option of adding more animated figures in the future.

So how can I make it a single FuncAnimation (or maybe something similar) that updates all three (potentially more) figures? Alternatively, what can I do to speed it up a bit?

#figure for current
amps = plt.figure(1)
ax1 = plt.subplot(xlim = (0,100), ylim = (0,500))
line, = ax1.plot([],[])
ax1.set_ylabel('Current (A)')

#figure for voltage
volts = plt.figure(2)
ax2 = plt.subplot()
rects1 = ax2.bar(ind1, voltV, width1)
ax2.grid(True)
ax2.set_ylim([0,6])
ax2.set_xlabel('Cell Number')
ax2.set_ylabel('Voltage (V)')
ax2.set_title('Real Time Voltage Data')
ax2.set_xticks(ind1)

#figure for temperature
temp = plt.figure(3)
ax3 = plt.subplot()
rects2 = ax3.bar(ind2, tempC, width2)
ax3.grid(True)
ax3.set_ylim([0,101])
ax3.set_xlabel('Sensor Number')
ax3.set_ylabel('temperature (C)')
ax3.set_title('Real Time Temperature Data')
ax3.set_xticks(ind2)

def updateAmps(frameNum):

    try:
    #error check for bad serial data
        serialString = serialData.readline()
        serialLine = [float(val) for val in serialString.split()]
        print (serialLine)

        if (len(serialLine) == 5):
            voltV[int(serialLine[1])] = serialLine[2]
            tempC[int(serialLine[3])] = serialLine[4]
            currentA.append(serialLine[0])
            if (len(currentA)>100):
                currentA.popleft()

        line.set_data(range(100), currentA)

    except ValueError as e:
    #graphs not updated for bad serial data
        print (e)

    return line,

#function to update real-time voltage data
def updateVolts(frameNum):

    for rects, h in zip(rects1,voltV):
        rects.set_height(h)

    return rects1

#function to update real-time temperature data
def updateTemp(frameNum):

    for rects, h in zip(rects2,tempC):
        rects.set_height(h)

    return rects2

Call to funcAnimation:

anim1 = animation.FuncAnimation(amps, updateAmps,
                                interval = 20, blit = True)
anim2 = animation.FuncAnimation(volts, updateVolts, interval = 25, blit = True)
anim3 = animation.FuncAnimation(temp, updateTemp, interval = 30, blit = True)

There are two mechanisms that can limit the speed here: (1) The time for reading the serial data and (2) The time to draw the artists to the canvas. It shouldn't be related to the number of FuncAnimations you have. I would suggest you investigate the two things seperately. Is it really much faster if you use a single Animation (e.g. only anim1)? — ImportanceOfBeingErnest
– ImportanceOfBeingErnest, Commented Mar 6, 2017 at 9:55
@ImportanceOfBeingErnest don't think it is the time for reading serial data. Having one FuncAnimation seems to speed things up a lot. It's a noticeable difference after hitting 3 FuncAnimations when the program starts to get a bit choppy. Also, I still need to add potentially another 3 live data charts which is why I need to find a way to optimize it now as I don't think it will go well if I follow the current implentation. — user3458571
– user3458571, Commented Mar 6, 2017 at 10:03
So the next test case would be to put all three plots as subplots to the same figure and use a single FuncAnimation. Compare this to the case of having three different animations. Is it significantly faster? If so, consider if having a single figure would actually be acceptable for your case. — ImportanceOfBeingErnest
– ImportanceOfBeingErnest, Commented Mar 6, 2017 at 10:11

Diziet Asahi · Accepted Answer · 2017-03-06 10:59:30Z

Echoing @ImportanceOfBeingErnest's comment, the obvious solution would be to use 3 subplots and only one FuncAnimation() call. You simply have to make sure your callback function returns a list of ALL artists to be updated at each iteration.

One drawback is that the update will happen a the same interval in all 3 subplots (contrary to the different timings you had in your example). You could potentially work around that by using global variables that count how many time the function has been called and only do some of the plots every so often for example.

#figure 
fig = plt.figure(1)
# subplot for current
ax1 = fig.add_subplot(131, xlim = (0,100), ylim = (0,500))
line, = ax1.plot([],[])
ax1.set_ylabel('Current (A)')

#subplot for voltage
ax2 = fig.add_subplot(132)
rects1 = ax2.bar(ind1, voltV, width1)
ax2.grid(True)
ax2.set_ylim([0,6])
ax2.set_xlabel('Cell Number')
ax2.set_ylabel('Voltage (V)')
ax2.set_title('Real Time Voltage Data')
ax2.set_xticks(ind1)

#subplot for temperature
ax3 = fig.add_subplot(133)
rects2 = ax3.bar(ind2, tempC, width2)
ax3.grid(True)
ax3.set_ylim([0,101])
ax3.set_xlabel('Sensor Number')
ax3.set_ylabel('temperature (C)')
ax3.set_title('Real Time Temperature Data')
ax3.set_xticks(ind2)

def updateAmps(frameNum):

    try:
    #error check for bad serial data
        serialString = serialData.readline()
        serialLine = [float(val) for val in serialString.split()]
        print (serialLine)

        if (len(serialLine) == 5):
            voltV[int(serialLine[1])] = serialLine[2]
            tempC[int(serialLine[3])] = serialLine[4]
            currentA.append(serialLine[0])
            if (len(currentA)>100):
                currentA.popleft()

        line.set_data(range(100), currentA)

    except ValueError as e:
    #graphs not updated for bad serial data
        print (e)

    return line,

#function to update real-time voltage data
def updateVolts(frameNum):

    for rects, h in zip(rects1,voltV):
        rects.set_height(h)

    return rects1

#function to update real-time temperature data
def updateTemp(frameNum):

    for rects, h in zip(rects2,tempC):
        rects.set_height(h)

    return rects2

def updateALL(frameNum):
    a = updateAmps(frameNum)
    b = updateVolts(frameNum)
    c = updateTemp(frameNum)
    return a+b+c

animALL = animation.FuncAnimation(fig, updateALL,
                                interval = 20, blit = True)

Thank you for the code above. You make a point about the update rates happening at the same interval. I do not fully understand the relationship between the frames and the interval. Could you elaborate with an example? Thank you.

Collectives™ on Stack Overflow

how to use FuncAnimation to update and animate multiple figures with matplotlib?

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related