Make scatterplot from arrays within a dataframe

Question

Let's say I have the following dataframe, for each month separately I have a bunch of data, stores in arrays for three variables :

ID         Y        X1        X2  month
0   [2,4,6,8] [2,4,6,8] [2,4,6,8]    01
1   [Nan,4,6,8] [1,3,5,4] [4,3,3,3]  02
2   [3,4,5,6] [1,9,7,7] [2,2,6,Nan]  03
3   [1,2,3,4] [5,6,7,8] [9,9,Nan,6]  04
4   [2,4,6,8] [2,4,6,8] [2,4,6,8]    05

What I ultimately want to do is to make a scatterplot between Y and X1 for month 01 with markers in darkblue, for month two in lightblue, and so on. Maybe I also want the scatterplot for Y and X2 in different shades of red as well in the same plot..
I tried this one:

df.iloc[0:1].plot.scatter(x = 'X1', y='Y')

but get the message that there are no numeric objects to plot...
Are the Nan values a problem???

Any ideas?! Thanks a lot for helping!

Answer 1

You need to change the structure of your data frame:

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np


data =  {"ID":[0,1,2,3,4],
         "Y":[np.array([2,4,6,8]), 
              np.array([np.nan,4,6,8]),
              np.array([3,4,5,6]), 
              np.array([1,2,3,4]), 
              np.array([2,4,6,8])],
        "X1":[np.array([2,4,6,8]), 
              np.array([1,2,5,4]),
              np.array([1,9,7,7]), 
              np.array([5,6,7,8]), 
              np.array([2,4,6,8])],
        "X2":[np.array([2,4,6,8]), 
              np.array([4,3,3,3]),
              np.array([2,2,6,np.nan]), 
              np.array([9,9,np.nan,6]), 
              np.array([2,4,6,8])],
        "month":[1,2,3,4,5]
}


df = pd.DataFrame(data)

check = 0
for v in range(len(df["Y"])):
    val_y = df["Y"][v]
    val_x1 = df["X1"][v]
    val_x2 = df["X2"][v]
    ID = df["ID"][v]
    month = df["month"][v]

    if check == 0:
        helper_dat = {"ID":ID,"Y":list(val_y),"X1":list(val_x1),"X2":list(val_x2),"month":month}
        new_df = pd.DataFrame(helper_dat)
    else:
        helper_dat = {"ID":ID,"Y":list(val_y),"X1":list(val_x1),"X2":list(val_x2),"month":month}
        helper = pd.DataFrame(helper_dat)
        new_df = new_df.append(helper,ignore_index=True)   
    check += 1

new_df now looks like this:

    ID    Y  X1   X2  month
0    0  2.0   2  2.0      1
1    0  4.0   4  4.0      1
2    0  6.0   6  6.0      1
3    0  8.0   8  8.0      1
4    1  NaN   1  4.0      2
5    1  4.0   2  3.0      2
6    1  6.0   5  3.0      2
7    1  8.0   4  3.0      2
8    2  3.0   1  2.0      3
9    2  4.0   9  2.0      3
10   2  5.0   7  6.0      3
11   2  6.0   7  NaN      3
12   3  1.0   5  9.0      4
13   3  2.0   6  9.0      4
14   3  3.0   7  NaN      4
15   3  4.0   8  6.0      4
16   4  2.0   2  2.0      5
17   4  4.0   4  4.0      5
18   4  6.0   6  6.0      5
19   4  8.0   8  8.0      5

now it is easy to plot the values:

plt.scatter(new_df["X1"],new_df["Y"],c=new_df["month"], marker='^',label="X1")
plt.scatter(new_df["X2"],new_df["Y"],c=new_df["month"], marker='o',label="X2")
plt.legend()

Edit: If you want to plot only one specific month:

plt.scatter(new_df[**new_df["month"]==4]["X1"]**,new_df[new_df["month"]==4]["Y"], marker='^',label="X1")
plt.scatter(new_df[new_df["month"]==4]["X2"],new_df[new_df["month"]==4]["Y"], marker='o',label="X2")

Found a way based on this Answer:

sc = plt.scatter(new_df["X1"],new_df["Y"],c=new_df["month"], marker='^',label="X1")
plt.scatter(new_df["X2"],new_df["Y"],c=new_df["month"], marker='o',label="X2")
lp = lambda i: plt.plot([],color=sc.cmap(sc.norm(i)),
                        label="Month {:g}".format(i))[0]
handles = [lp(i) for i in np.unique(new_df["month"])]
plt.legend(handles=handles,bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
plt.show()