python condensing repetitive code and subplots

Question

how do I condense these lines of code and say, if I want to plot all 3 graphs separately, how do I use the matplotlib subplot function to do it? below is what I've done but am unsure of how to condense it:

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
sns.set()

totalpop = 18000
subpop = [300, 400, 500]
samplesize_list1 = [10, 20, 30] #sample size 60
samplesize_list2 = [40, 50, 60] #sample size 150
samplesize_list3 = [70, 80, 90] #sample size 240
label_list = ['size60', 'size150', 'size240']
std_list = [300, 500, 700]
mean_list = [450, 670, 780]
repeat = 500 #repeated random sampling of 500 sampling outcomes

samplingdist1 = []
for i in range(500):
    sample1 = []
    for i in range(len(samplesize_list1)):
        for j in range(samplesize_list1[i]):
            s = 0
            while True:
                s = np.random.normal(mean_list[i], std_list[i], 1).tolist()
                if s[0] > 0: 
                    break
            sample1 += s
    samplingdist1.append(np.median(sample1))
sns.distplot(samplingdist1, label = 'size60')

samplingdist2 = []
for i in range(500):
    sample2 = []
    for i in range(len(samplesize_list2)):
        for j in range(samplesize_list2[i]):
            s = 0
            while True:
                s = np.random.normal(mean_list[i], std_list[i], 1).tolist()
                if s[0] > 0:
                    break
            sample2 += s
    samplingdist2.append(np.median(sample2))
sns.distplot(samplingdist2, label = 'size150')

samplingdist3 = []
for i in range(500):
    sample3 = []
    for i in range(len(samplesize_list3)):
        for j in range(samplesize_list3[i]):
            s = 0
            while True:
                s = np.random.normal(mean_list[i], std_list[i], 1).tolist()
                if s[0] > 0:
                    break
            sample3 += s
    samplingdist3.append(np.median(sample3))
sns.distplot(samplingdist3, label = 'size240')

the graphs I got from this were plotted in a single graph, say I want to plot them separately and label it 'size60' 'size150' 'size240' respectively. how do I approach this?

score 0 · Answer 1 · answered Sep 21 '18 at 10:19

Here is a condensed solution for you with brief explanations added as comments

# Put your imports here

totalpop = 18000
subpop = [300, 400, 500]
samplesize_list = [[10, 20, 30], [40, 50, 60],[70, 80, 90]]  # 3 lists combined into one
# Your label, std, mean and repeat data here

fig = plt.figure(figsize=(13, 3))
axes = fig.subplots(nrows=1, ncols=3) # Create a figure with 3 columns and 1 row

for ind, ax in enumerate(axes.flatten()): # Enumerate here to access index for samplesize
    samplingdist = []
    for i in range(500):
        sample = []
        for i in range(len(samplesize_list[ind])): # ind accesses the corresponding sublist
            for j in range(samplesize_list[ind][i]):
                s = 0
                while True:
                    s = np.random.normal(mean_list[i], std_list[i], 1).tolist()
                    if s[0] > 0: 
                        break
                sample += s
        samplingdist.append(np.median(sample))
    sns.distplot(samplingdist, label = label_list[ind], ax=ax) # Pass the subplot ax for plotting
    ax.legend()  # Show the legend

Output

python condensing repetitive code and subplots

1 Answers1