nested loops resulting in unwanted repeated figures

Question

I know this has been asked before but I cannot figure out how to word my question to find what I am looking for. I am still fairly new to coding and am trying to generate plots using nested for loops. I intend to create 4 plots, but each plot is created 4 times, resulting in 16 figures total. What is causing the repetition?

I've already checked plotting multiple plots generated inside a for loop on the same axes python but this was irrelevant.

Here is my code:

###(this is sample data in case you are trying to recreate the code)
df24avg=pd.dataframe({'pm10_3135_2018':[30,34,32,44,45,46,59,54,59,30],
'nox_3135(ppb)':[20,29,27,31,33,14,34,23,32,31],
'CO_3135(ppm)':[0.8,0.9,0.1,0.2,0.5,0.5,0.7,0.9,0,9,0.3],
'O3_mda8_3135':[42,45,47,51,52,52,57,67,65,70],
'pm25_3135_2018':[6,7,6,7,4,5,2,11,9,18]})

x = [df24avg.pm10_3135_2018,df24avg['nox_3135(ppb)'],df24avg['CO_3135(ppm)'],df24avg.O3_mda8_3135]
y = df24avg.pm25_3135_2018
xlab = ["PM10 (ug/m^3)", "NOx (ppb)", "CO (ppm)", "O3 MDA8 (ppb)"]

for xcol in x:
    for lab in xlab:

        fig, ax = plt.subplots()
        ax = plt.gca()
        ax.plot(xcol, y, color='xkcd:red',linestyle='None',marker='o')
        ax.set_xlabel(lab,fontsize=15)
        ax.set_ylabel('PM2.5 (ug/m^3)',fontsize=15)
        ax.set_ylim(0,)
        ax.set_xlim(0,)
        #ax.set_title("{} vs {}".format(x1, ylab))
    
        #ax.legend(fontsize='medium', loc='lower right')
        fig.tight_layout()

I also would like to save each figure with different names of my choosing but am unsure how to do that in the loop without creating a third loop. I am also not opposed to creating another loop just want to avoid this issue of figures repeating. Thank you!

first guess is the double loop. Can you explain in clear english what the criteria is for splitting your dataframe into four parts to plot? — Umar.H, Apr 26 '21 at 18:03
You are defining your figure within the second loop so that's 16 figures. You _probably_ want to define the figure in your first loop, use the second loop to plot the 4 lines and then set the axis properties outside of the second loop (again since there's only one plot...). Also probably need to reverse the order of your loops. — ALollz, Apr 26 '21 at 18:05
IIUC, one loop: `for xcol, lab in zip(x, xlab):`. Also, `ax = plt.gca()` can be eliminated. If you want specific filenames, create another list and add that to the `zip`. — BigBen, Apr 26 '21 at 18:06

score 2 · Accepted Answer · answered Apr 26 '21 at 18:14

If I understand correctly, you need one loop with zip:

for xcol, lab in zip(x, xlab):
    fig, ax = plt.subplots()
    ax.plot(xcol, y, color='xkcd:red',linestyle='None',marker='o')

ax = plt.gca() should be removed as ax is already determined in the subplots call.

If you have a list of unique filenames for each plot, simply add that to zip:

fnames = ['foo.png', 'bar.png', 'baz.png', 'grr.png']

for xcol, lab, fname in zip(x, xlab, fnames):
    ...
    plt.savefig(fname)

score 0 · Answer 2 · answered Apr 26 '21 at 18:25

You can use range(len()) function instead of directly using values.

for i in range(len(x)):
#    for lab in xlab: # you don't need this line
        fig, ax = plt.subplots()
        ax = plt.gca()
        ax.plot(x[i], y, color='xkcd:red',linestyle='None',marker='o') # Modified with index
        ax.set_xlabel(xlab[i],fontsize=15) # Also used index
        ax.set_ylabel('PM2.5 (ug/m^3)',fontsize=15)
        ax.set_ylim(0,)
        ax.set_xlim(0,)
        #ax.set_title("{} vs {}".format(x1, ylab))
    
        #ax.legend(fontsize='medium', loc='lower right')
        fig.tight_layout()

nested loops resulting in unwanted repeated figures

2 Answers2