[FIXED] Error with plt.savefig in the loop

Issue

At first, I have to say that I'm a true beginner in Python (and in programming itself) so this may be a silly question but I couldn't find a solution.

I load data from 3 different .csv files then make some calculations and in the end i want to save 3 heatmaps. My problem is with saving. When i use plt.savefig('whatever.jpg') it obviously saves only last heatmap in the loop. I tried to change it to plt.savefig('{}.jpg'.format(i)) but I got an error and I don't know how to fix. I'll appreciate any hints! (Python 3 from Anaconda)

import seaborn as sns
import pandas as pd 
import matplotlib.pyplot as plt
%matplotlib inline

instrumenty = ['gold','sp500','dax']

for i in instrumenty:

    i = pd.read_csv(i+'_m.csv', sep=',')
    i['Miesiąc'] = pd.DatetimeIndex(i['Data']).month
    i['Rok'] = pd.DatetimeIndex(i['Data']).year
    i['Zmiana']=i['Zamkniecie'].pct_change()
    i_pt = i[1:].pivot_table(index='Rok',columns='Miesiąc',values='Zmiana')
    sns.heatmap(i_pt, linewidths=1, cmap='PiYG')
    plt.savefig('{}.png'.format(i))
    plt.clf()

And the error:

   FileNotFoundError                         Traceback (most recent call last)
<ipython-input-30-5729b6a0480f> in <module>()
     14     i_pt = i[1:].pivot_table(index='Rok',columns='Miesiąc',values='Zmiana')
     15     sns.heatmap(i_pt, linewidths=1, cmap='PiYG')
---> 16     plt.savefig('{}.png'.format(i))
     17     plt.clf()
     18 

C:\Users\user\Anaconda3\lib\site-packages\matplotlib\pyplot.py in savefig(*args, **kwargs)
    694 def savefig(*args, **kwargs):
    695     fig = gcf()
--> 696     res = fig.savefig(*args, **kwargs)
    697     fig.canvas.draw_idle()   # need this if 'transparent=True' to reset colors
    698     return res

C:\Users\user\Anaconda3\lib\site-packages\matplotlib\figure.py in savefig(self, *args, **kwargs)
   1561             self.set_frameon(frameon)
   1562 
-> 1563         self.canvas.print_figure(*args, **kwargs)
   1564 
   1565         if frameon:

C:\Users\user\Anaconda3\lib\site-packages\matplotlib\backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
   2230                 orientation=orientation,
   2231                 bbox_inches_restore=_bbox_inches_restore,
-> 2232                 **kwargs)
   2233         finally:
   2234             if bbox_inches and restore_bbox:

C:\Users\user\Anaconda3\lib\site-packages\matplotlib\backends\backend_agg.py in print_png(self, filename_or_obj, *args, **kwargs)
    530         renderer.dpi = self.figure.dpi
    531         if is_string_like(filename_or_obj):
--> 532             filename_or_obj = open(filename_or_obj, 'wb')
    533             close = True
    534         else:

Even without plt.savefig i am doing something wrong because the result is one strange chart with 3 legends instead of 3 charts.

Solution

You override i here:

instrumenty = ['gold','sp500','dax']
for i in instrumenty:
    i = pd.read_csv(i+'_m.csv', sep=',')

Better use name:

for name in instrumenty:
    # use `name`
    i = pd.read_csv(name +'_m.csv', sep=',')
    i['Miesiąc'] = pd.DatetimeIndex(i['Data']).month
    i['Rok'] = pd.DatetimeIndex(i['Data']).year
    i['Zmiana']=i['Zamkniecie'].pct_change()
    i_pt = i[1:].pivot_table(index='Rok',columns='Miesiąc',values='Zmiana')
    sns.heatmap(i_pt, linewidths=1, cmap='PiYG')
    plt.savefig('{}.png'.format(name))
    plt.clf()

This is the important part:

plt.savefig('{}.png'.format(name))
plt.clf()

Answered By - Mike Müller