In [1]:

    

%matplotlib inline
import pandas as pd
import matplotlib.pyplot as plt
import csv
import numpy as np

names = ["date", "city", "state", "country", "shape", "duration_seconds", "duration_reported", "description", "report_date", "latitude", "longitude"]

ufo = pd.read_csv("/srv/nbgrader/data/ufo-scrubbed-geocoded-time-standardized.csv", names = names, parse_dates = ["date", "report_date"])


    

In [2]:

    

total_time = ufo.groupby(["state"])["duration_seconds"].sum()

total_sightings = ufo.groupby(["state"])["report_date"].count()


    

In [3]:

    

df = pd.DataFrame([total_time, total_sightings])


    

In [4]:

    

df


    

    
Out[4]:







  

    

      
state
      
ab
      
ak
      
al
      
ar
      
az
      
bc
      
ca
      
co
      
ct
      
dc
      
...
      
tx
      
ut
      
va
      
vt
      
wa
      
wi
      
wv
      
wy
      
yk
      
yt
    
  
  

    

      
duration_seconds
      
590526.0
      
1498068.0
      
962845.5
      
67177514.5
      
15990939.1
      
869341.82
      
37928460.47
      
4551714.1
      
12670360.05
      
114961.23
      
...
      
8719429.25
      
3521102.0
      
13738539.5
      
330819.5
      
57812602.14
      
2729937.3
      
3032327.0
      
377534.001
      
10218.0
      
15974.0
    
    

      
report_date
      
333.0
      
354.0
      
691.0
      
666.0
      
2689.0
      
788.00
      
9655.00
      
1505.0
      
968.00
      
99.00
      
...
      
3677.00
      
743.0
      
1393.0
      
307.0
      
4268.00
      
1333.0
      
486.0
      
205.000
      
7.0
      
13.0
    
  

2 rows × 67 columns

In [5]:

    

df.rename({'report_date': 'total_sightings'}, inplace=True)


    

In [6]:

    

ufo_vals = df.transpose()


    

In [7]:

    

ufo_vals.describe()


    

    
Out[7]:







  

    

      

      
duration_seconds
      
total_sightings
    
  
  

    

      
count
      
6.700000e+01
      
67.000000
    
    

      
mean
      
7.361724e+06
      
1112.462687
    
    

      
std
      
1.631502e+07
      
1462.977146
    
    

      
min
      
1.021800e+04
      
7.000000
    
    

      
25%
      
4.041390e+05
      
200.500000
    
    

      
50%
      
1.781951e+06
      
691.000000
    
    

      
75%
      
5.676579e+06
      
1389.500000
    
    

      
max
      
8.465105e+07
      
9655.000000
    
  

In [8]:

    

ufo_vals.loc["mi"]


    

    
Out[8]:





duration_seconds    8046018.6
total_sightings        2071.0
Name: mi, dtype: float64

In [9]:

    

import cartopy


    

In [10]:

    

states = cartopy.io.shapereader.natural_earth(resolution='110m', category='cultural',
                                    name='admin_1_states_provinces_lakes_shp')


    

In [11]:

    

reader =cartopy.io.shapereader.Reader(states)


    

In [12]:

    

all_states = list(reader.records())


    

In [13]:

    

all_states[0]


    

    
Out[13]:





<Record: <shapely.geometry.multipolygon.MultiPolygon object at 0x7fdfb2cdf1d0>, {'scalerank': 2, 'featurecla': 'Admin-1 scale rank', 'adm1_code': 'USA-3514', 'diss_me': 3514, 'adm1_cod_1': 'USA-3514', 'iso_3166_2': 'US-MN', 'wikipedia': 'http://en.wikipedia.org/wiki/Minnesota', 'sr_sov_a3': 'US1', 'sr_adm0_a3': 'USA', 'iso_a2': 'US', 'adm0_sr': 1, 'admin0_lab': 2, 'name': 'Minnesota', 'name_alt': 'MN|Minn.', 'name_local': '', 'type': 'State', 'type_en': 'State', 'code_local': 'US32', 'code_hasc': 'US.MN', 'note': '', 'hasc_maybe': '', 'region': 'Midwest', 'region_cod': '', 'region_big': 'West North Central', 'big_code': '', 'provnum_ne': 0, 'gadm_level': 1, 'check_me': 10, 'scaleran_1': 2, 'datarank': 1, 'abbrev': 'Minn.', 'postal': 'MN', 'area_sqkm': 0.0, 'sameascity': -99, 'labelrank': 0, 'featurec_1': 'Admin-1 scale rank', 'admin': 'United States of America', 'name_len': 9, 'mapcolor9': 1, 'mapcolor13': 1}, <fields>>

In [14]:

    

_.geometry


    

    
Out[14]:

In [15]:

    

geometries = [state.geometry for state in all_states]


    

In [16]:

    

fig = plt.figure()
ax = fig.add_subplot(111, projection = cartopy.crs.LambertConformal())
ax.set_extent([-125, -66.5, 20, 50], cartopy.crs.Geodetic())
ax.add_geometries(geometries, crs=cartopy.crs.PlateCarree(), edgecolors='k', facecolors = 'k')


    

    
Out[16]:





<cartopy.mpl.feature_artist.FeatureArtist at 0x7fdfa726aba8>






    

In [17]:

    

import matplotlib.cm as cm


    

In [18]:

    

cm.viridis(0.4)


    

    
Out[18]:





(0.16362499999999999, 0.47113300000000002, 0.55814799999999998, 1.0)

In [19]:

    

import ipywidgets


    

In [20]:

    

@ipywidgets.interact(field = ["total_sightings", "duration_seconds"] )
def make_plot(field):
    fig = plt.figure(figsize=(10,10))
    ax = fig.add_subplot(111, projection = cartopy.crs.LambertConformal())
    ax.set_extent([-125, -66.5, 20, 50], cartopy.crs.Geodetic())
    colors = {'IL': 'k', 'CA': 'r'}
    min_val = np.log10(ufo_vals[field].min())
    max_val = np.log10(ufo_vals[field].max())
    for state in all_states:
        postal_code = state.attributes['postal'].lower()
        n = np.log10(ufo_vals.loc[postal_code][field])
        norm = (n - min_val)/(max_val - min_val)
        color = cm.viridis(norm)
        ax.add_geometries( [state.geometry], crs = cartopy.crs.PlateCarree(),
                         facecolors = [color])


    

In [21]:

    

mass = ufo[ ufo["state"] == "ma" ]


    

In [22]:

    

import numpy as np


    

In [23]:

    

fig = plt.figure(figsize=(10,10))
ax = fig.add_subplot(111, projection = cartopy.crs.LambertConformal())
ax.set_extent([-75, -70, 40, 50], cartopy.crs.Geodetic())
ax.add_geometries(geometries, crs = cartopy.crs.PlateCarree(), zorder=0)
ax.scatter(mass["longitude"], mass["latitude"], transform = cartopy.crs.Geodetic(),
          c=np.log10(mass["duration_seconds"]))


    

    
Out[23]:





<matplotlib.collections.PathCollection at 0x7fdfa70582b0>






    

In [24]:

    

import cartopy.io.img_tiles as cimgt


    

In [25]:

    

fig = plt.figure(figsize=(10,10))
ax = fig.add_subplot(111, projection = cartopy.crs.GOOGLE_MERCATOR)
ax.set_extent([-75, -70, 40, 50], cartopy.crs.Geodetic())
ax.add_image(cimgt.GoogleTiles(), 5)
ax.scatter(mass["longitude"], mass["latitude"], transform = cartopy.crs.Geodetic(),
          c=np.log10(mass["duration_seconds"]))


    

    
Out[25]:





<matplotlib.collections.PathCollection at 0x7fdfa6fe1a90>






    

In [26]:

    

fig = plt.figure()
ax = fig.add_subplot(111, projection = cartopy.crs.Mollweide())
ax.coastlines()


    

    
Out[26]:





<cartopy.mpl.feature_artist.FeatureArtist at 0x7fdfa50704a8>