In [3]:

    

#Importing necessary libraries
import pandas as pd
import numpy as np 
import matplotlib.pyplot as plt
plt.style.use('ggplot')
import gmplot
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
%matplotlib inline


    

In [4]:

    

# Read Data from csv file
df = pd.read_csv("part3_data.csv")
df.describe()


    

    
Out[4]:






  

    

      

      
dates
      
latitude
      
longitude
      
quant1
      
quant2
      
quant3
    
  
  

    

      
count
      
1.000000e+04
      
10000.000000
      
10000.000000
      
10000.000000
      
10000.000000
      
10000.000000
    
    

      
mean
      
1.467361e+09
      
1.026321
      
-0.082380
      
1.737077
      
0.031580
      
4980.333493
    
    

      
std
      
9.141892e+06
      
51.545570
      
103.546299
      
1.012192
      
17.456411
      
2889.413452
    
    

      
min
      
1.451612e+09
      
-89.995631
      
-179.756945
      
0.001356
      
-29.979777
      
0.723583
    
    

      
25%
      
1.459531e+09
      
-43.319612
      
-89.584439
      
0.854677
      
-15.223848
      
2490.176566
    
    

      
50%
      
1.467432e+09
      
1.219217
      
0.070930
      
1.726674
      
-0.234758
      
4920.907779
    
    

      
75%
      
1.475310e+09
      
46.321107
      
88.911360
      
2.618524
      
15.397590
      
7499.437028
    
    

      
max
      
1.483138e+09
      
89.974703
      
179.946696
      
3.499813
      
29.994035
      
9999.376454
    
  

In [5]:

    

def analysis(df):
    df.describe()
    df.mean()
    plt.rcParams['figure.figsize'] = (15,15)
    plt.scatter(df.latitude, df.longitude,c=np.arange(len(df.latitude)))
    plt.title("Spatial distribution of coordinates")
    plt.xlabel("Latitude")
    plt.ylabel("Longitude")
    plt.show()
    gmap = gmplot.GoogleMapPlotter(40.088, -88.281, 16)
    # Scatter plot on Google Map
    gmap.scatter(df["latitude"],df["longitude"], size=40, marker=False)
    gmap.draw("scatter.html")
    #Heatmap on Google Map
    gmap.heatmap(df["latitude"],df["longitude"])
    gmap.draw("heatmap.html")
    #<--- --->
    cat_uniq = df.categorical.unique()
    quants = ['quant1', 'quant2', 'quant3']
    color=np.random.rand(10,3)
    # Subplots 
    for i in range(1,4):
        plt.subplot(2,2,i)    
        plt.bar(np.arange(len(cat_uniq)), df.groupby('categorical').sum()[quants[i-1]], color=color)
        plt.xticks(np.arange(len(cat_uniq)), cat_uniq, rotation = 45)
        plt.title("Sum of "+ quants[i-1] + " across all categories")
    plt.subplots_adjust(wspace=.2, hspace=.3)
    plt.show()
    ax = Axes3D(plt.gcf())
    ax.scatter(df['quant1'].tolist(),df['quant2'].tolist(),df['quant3'].tolist(), c=np.arange(len(df.latitude)))
    ax.set_xlabel('Quant 1')
    ax.set_ylabel('Quant 2')
    ax.set_zlabel('Quant 3')
    ax.set_title('3D distribution of all the quant values')


    

In [6]:

    

analysis(df)


    

In [ ]:

    




    

In [ ]: