URL ='http://www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELPRDC5087258'

DEBUG = False

def __init__(self):

#Parsing local CSV.

location = r'/Desktop/WorkingWithOpenData/FarmersMarkets.csv'

df = pd.read_csv(location)

df = pd.DataFrame.from_csv(location)

df.columns

def sort_top_states(df):

#df -> DataFrame

#Find the top5 states with the most farmers markets.

Sorted = df.sort(['State'], ascending=True)

top_states = dict

for states in Sorted.State:

if states not in top_states:

top_states[states] = 1

else:

top_states[states] += 1

sorted_top_five = dict(sorted(top_states.iteritems(), key=operator.itemgetter(1), reverse=True)[:5])

sorted_bottom_five = dict(sorted(top_states.iteritems(), key=operator.itemgetter(1), reverse=False)[:8])

print "Top Five States :\n", sorted_top_five

print "\nBottom Five States :\n", sorted_bottom_five

#Incorrectly Spelled State Names...

#Way to go Department of Agriculture, remove them.

del sorted_bottom_five['Calafornia']

del sorted_bottom_five['Virigina']

del sorted_bottom_five['Miinesota']

def plot_top_states(df):

#iPython Command

%matplotlib inline

"""

Matplotlib Top 5 Farmers Market States

"""

states_label = sorted_top_five.keys()

y_pos = np.arange(len(states_label))

#error = np.random.rand(len(people))

number_of_markets = sorted_top_five.values()

plt.barh(y_pos, number_of_markets, align='center', alpha=0.4)

plt.yticks(y_pos, states_label)

plt.xlabel('Farmers Markets per State')

plt.title('Top Five Farmers Market States')

plt.show()

def sort_top_counties(df):

#Sort DataFrame by Top Counties.

County_Sorted = df.sort(['County'], ascending=True)

top_counties = dict()

for counties in County_Sorted.County:

if counties not in top_counties:

top_counties[counties] = 1

else:

top_counties[counties] += 1

all_counties = dict(sorted(top_counties.iteritems(), key=operator.itemgetter(1), reverse=True))

counties_top_five = dict(sorted(top_counties.iteritems(), key=operator.itemgetter(1), reverse=True)[:10])

counties_bottom_five = dict(sorted(top_counties.iteritems(), key=operator.itemgetter(1), reverse=False)[:10])

#All of the included Farmers Market Counties

if(DEBUG):

all_counties.keys()

print(len(all_counties)) #1423 Counties

print "Top Counties with Farmers Markets \n", counties_top_five

print "\nCounties with Least Farmers Markets \n", counties_bottom_five

#Drop all NaN Values from County Field

County_Sorted = County_Sorted[pd.notnull(County_Sorted['County'])]

columns = ["County"]

organized = pd.DataFrame(County_Sorted, columns=columns)

organized.head(20) #Outputs first 20 entries pointing to Head

def obesity_rates(df):

#Obesity Rates in Los Angeles County and Farmers Market correlation.

#http://county-food.findthebest.com/l/3344/Los-Angeles

LA_obesity_rate = 21.5

LA_markets = counties_top_five['Los Angeles'] #116 total farmers markets in Los Angeles

Alameda_markets = all_counties['ALAMEDA'] #2 total farmers markets

SD_markets = all_counties['San Diego'] #47

Orange_markets = all_counties['Orange'] #60

Riverside_markets = all_counties['Riverside'] #18

SB_markets = all_counties['San Bernardino'] #22

CA_total_markets = sorted_top_five['California'] #754 total farmers markets in CA

total_farmers_markets = len(County_Sorted) #7,390 Total Farmers Markets

LA_percent = (float(LA_markets)/float(CA_total_markets)) * 100 #15.4

Alameda_percent = (float(Alameda_markets)/float(CA_total_markets)) * 100

SD_percent = (float(SD_markets)/float(CA_total_markets)) * 100

Orange_percent = (float(Orange_markets)/float(CA_total_markets)) * 100

Riverside_percent = (float(Riverside_markets)/float(CA_total_markets)) * 100

SB_percent = (float(SB_markets)/float(CA_total_markets)) * 100

percent_dict = {}

percent_dict['Los Angeles'] = LA_percent

percent_dict['Alameda'] = Alameda_percent

percent_dict['San Diego'] = SD_percent

percent_dict['Orange'] = Orange_percent

percent_dict['Riverside'] = Riverside_percent

percent_dict['San Bernardino'] = SB_percent

print "Total LA County Farmers Markets: ", total_farmers_markets

def obesity_top_counties(df):

"""

Matplotlib Top 6 CA Counties and Farmers Market Percentage

"""

#fig = plt.figure()

%matplotlib inline

percent_label = percent_dict.keys()

y_pos = np.arange(len(percent_label))

#error = np.random.rand(len(people))

number_of_markets = percent_dict.values()

plt.barh(y_pos, number_of_markets, align='center', alpha=0.4)

plt.yticks(y_pos, percent_label)

plt.xlabel('County Percentage Per State')

plt.title('Top Six California Counties')

#Add Same bar graph but obesity rates in counties.

obesity_dict = {}

obesity_dict['Los Angeles'] = 21.5

obesity_dict['Alameda'] = 23.0

obesity_dict['Orange'] = 22.0

obesity_dict['San Diego'] = 24.0

obesity_dict['Riverside'] = 27.0

obesity_dict['San Bernardino'] = 28.0

obesity_label = obesity_dict.keys()

x_pos = np.arange(len(obesity_label))

obesity_len = obesity_dict.values()

plt.barh(x_pos, obesity_len, align='center', alpha=0.4)

plt.yticks(x_pos, obesity_label)

plt.xlabel('County Percentage Per State')

plt.title('Top Six California Counties')

#plt.legend( (number_of_markets, obesity_len), ("Obesity", "Density"))

plt.show()

def fast_food_per_state(df):

#Number of Fast Food Restaurants Per State

#http://www.businessinsider.com/this-interactive-map-shows-exactly-how-many-fast-food-restaurants-there-are-in-every-state-2012-1

CA_fast_food = 27120

NY_fast_food = 17461

MI_fast_food = 7005

OH_fast_food = 7137

IL_fast_food = 8790

N = 5

farmers_markets = (sorted_top_five['California'], sorted_top_five['New York'], sorted_top_five['Michigan'],

sorted_top_five['Ohio'], sorted_top_five['Illinois'])

ind = np.arange(N) # the x locations for the groups

width = 0.35 # the width of the bars

fig, ax = plt.subplots()

rects1 = ax.bar(ind, farmers_markets, width, color='r')

fast_food = [CA_fast_food, NY_fast_food, MI_fast_food, OH_fast_food, IL_fast_food]

rects2 = ax.bar(ind+width, fast_food, width, color='y')

ax.set_ylabel('Total')

ax.set_title('Fast Food Vs. Farmers Markets')

ax.set_xticks(ind+width)

x = ("California", "New York", "Michigan", "Ohio", "Illinois")

ax.set_xticklabels( x )

ax.legend( (rects1[0], rects2[0]), ('FarmersMarkets', 'FastFood') )

def autolabel(rects):

# attach some text labels

for rect in rects:

height = rect.get_height()

ax.text(rect.get_x()+rect.get_width()/2., 1.05*height, '%d'%int(height),

ha='center', va='bottom')

autolabel(rects1)

autolabel(rects2)

plt.show()

def final_summary(df):

"""

Percentage of Farmers Markets Per 100 Fast Food Joints

"""

ca = float((sorted_top_five['California']) / float(CA_fast_food)) * 100

ny = float((sorted_top_five['New York']) / float(NY_fast_food)) * 100

mi = float((sorted_top_five['Michigan']) / float(MI_fast_food)) * 100

oh = float((sorted_top_five['Ohio']) / float(OH_fast_food)) * 100

il = float((sorted_top_five['Illinois']) / float(IL_fast_food)) * 100

percentage_df = pd.DataFrame([ca,ny,mi,oh,il], columns=["% of Farmers Markets per 100 Fast Food"], index=['CA', 'NY', 'MI', 'OH', 'IL'])

percentage_df

"""

RESULTS:

California -> 35:1 Ratio (Fast Food / Farmers Markets)

New York -> 27:1 Ratio (Fast Food / Farmers Markets)

Michigan -> 22:1 Ratio (Fast Food / Farmers Markets)

Ohio -> 24:1 (Fast Food / Farmers Markets)

Illinois -> 26:1 (Fast Food / Farmers Markets)