I am creating multiple subplots from a dataframe containing reaction results (70 per dataframe) and I want to plot the reactions 12 by 12 to have a quick look before going through a more thorough analysis. Since 70/12 leaves a remainder, a simple implementation would run out of bound. I can solve this by using a "if,else" statement but it is not really elegant nor efficient. I would like to know if there was a nicer option. warDf is of size 70, meanDf is of size 130x70. time, pcmean and ncmean are of size 130. I am using the libraries pandas (pd), numpy (np) and matplotlib.pyplot (plt),
it=int(np.ceil(np.size(warDf)/12))# defining what to loop over
kk=0
for kk in np.arange(0,it):
#declaring the subplots
fig,axes=plt.subplots(nrows=3,ncols=4,sharex='col',sharey='row')
#transforming axes in a usable list
axe_list=[item for sublist in axes for item in sublist]
# checking that I don't run out of bond
if (12*kk+12<np.size(warDf)):
k=0
# plotting each graph in its corresponding subplot
for k in np.arange(0,12):
ax=axe_list.pop(0)
ax.plot(time,meanDf.iloc[:,12*kk+k],label=(meanDf.columns[12*kk+k]),color='blue')
ax.plot(time,pcmean,color='green')
ax.plot(time,ncmean,color='red')
ax.set_ylabel('fluorescence')
ax.set_xlabel('time/ (s)')
ax.legend()
else: # better option??
k=0
s2=np.size(warDf)-12*kk
for k in np.arange(0,s2):
ax=axe_list.pop(0)
ax.plot(time,meanDf.iloc[:,12*kk+k],label=(meanDf.columns[12*kk+k]),color='blue')
ax.plot(time,pcmean,color='green')
ax.plot(time,ncmean,color='red')
ax.set_ylabel('fluorescence')
ax.set_xlabel('time/ (s)')
ax.legend()
plt.show()
You could use the min()
function. Replace the entire if/else
with this:
k=0 # note: you don't have to pre-define k here
s2 = min(np.size(warDf) - 12 * kk, 12) # new part
for k in np.arange(0,s2): # the rest is the same as in the body of the else
ax=axe_list.pop(0)
ax.plot(time,meanDf.iloc[:,12*kk+k],label=(meanDf.columns[12*kk+k]),color='blue')
ax.plot(time,pcmean,color='green')
ax.plot(time,ncmean,color='red')
ax.set_ylabel('fluorescence')
ax.set_xlabel('time/ (s)')
ax.legend()
You currently have
if (12 * kk + 12 < np.size(warDf)):
s2 = 12 # define s2 as a variable here as well
for k in np.arange(0, s2):
# ...
else:
s2 = np.size(warDf) - 12 * kk
for k in np.arrange(0, s2):
# ...
Rearranging that first if
, we can get:
if (12 < np.size(warDf) - 12 * kk):
s2 = 12
# ...
else:
s2 = np.size(warDf) - 12 * kk
# ...
Now you can see that the right side of the if and the assignment to s2 are the same. If 12 is less, use 12. Otherwise, use np.size(warDf) - 12 * kk
. This is the definition of min()
.