Sign in for more options. Each user starts with 1,000 free credits.
UKO5A
In [7]:
#CrossCompute
target_folder = '/tmp'
ethnicity_select = """
Hispanic
Black Non-Hispanic
Not Stated/Unknown
Other Race/ Ethnicity
Hispanic
Asian and Pacific Islander
White Non-Hispanic
"""
In [8]:
race = ethnicity_select.strip().splitlines()[0]
In [9]:
import pandas as pd
# data = pd.read_csv('https://data.cityofnewyork.us/api/views/jb7j-dtam/rows.csv?accessType=DOWNLOAD
#using pandas to manipulate the data
data = pd.read_csv('https://data.cityofnewyork.us/api/views/jb7j-dtam/rows.csv?accessType=DOWNLOAD', na_values='.')
data['Race Ethnicity'].unique()
data_frame = data[(data['Race Ethnicity'] == race)]
In [10]:
data.head()
Out[10]:
In [13]:
data.dtypes
Out[13]:
In [14]:
import numpy as np
data.replace('.', np.nan)
##fixing the data set since it was not returning the right data type.
Out[14]:
In [15]:
data.iloc[1]
Out[15]:
In [46]:
data.dtypes
Out[46]:
In [16]:
%matplotlib inline
In [48]:
data['Deaths'] = pd.to_numeric(data['Deaths'], errors='coerce')
data['Death Rate'] = pd.to_numeric(data['Death Rate'], errors='coerce')
data['Age Adjusted Death Rate'] = pd.to_numeric(data['Age Adjusted Death Rate'], errors='coerce')
data['Sex'] = data['Sex'].astype('category')
data['Race Ethnicity'] = data['Race Ethnicity'].astype('category')
data['Leading Cause'] = data['Leading Cause'].astype('category')
In [17]:
data.columns
Out[17]:
In [18]:
data.dtypes
Out[18]:
In [19]:
selected_t = data_frame[data_frame['Year'] == 2010]
selected_t[:5]
Out[19]:
In [21]:
data['Deaths'] = pd.to_numeric(data['Deaths'], errors='coerce')
data['Death Rate'] = pd.to_numeric(data['Death Rate'], errors='coerce')
data['Age Adjusted Death Rate'] = pd.to_numeric(data['Age Adjusted Death Rate'], errors='coerce')
data['Sex'] = data['Sex'].astype('category')
data['Race Ethnicity'] = data['Race Ethnicity'].astype('category')
data['Leading Cause'] = data['Leading Cause'].astype('category')
In [23]:
g = selected_t.groupby('Leading Cause')
In [25]:
g['Death Rate'].mean()
Out[25]:
In [27]:
%matplotlib inline
x_plot = g['Deaths'].sum().plot.barh(figsize=(20, 5))
In [29]:
fig = x_plot.get_figure()
fig
Out[29]:
In [30]:
target_path = target_folder + '/x.png'
fig.savefig(target_path)
print('abc_image_path = %s' % target_path)
In [31]:
data.iloc[0]
Out[31]:
In [34]:
import pandas as pd
In [35]:
data_frame['Leading Cause'].unique()
Out[35]:
In [33]:
%matplotlib inline