-
Notifications
You must be signed in to change notification settings - Fork 1
/
final_doc.py
507 lines (405 loc) · 18.6 KB
/
final_doc.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
# UW NetID (Nanda Sundaresan): nandas
# UW NetID (Vineeth Sai Narajala): vineeth7
# CSE 160
# Homework 7: Final project
import csv
from operator import itemgetter
import matplotlib.pyplot as plt
import pylab
import pandas as pd
import seaborn as sns
plt.rcParams["figure.figsize"] = (16, 8)
pylab.rcParams["figure.figsize"] = (16, 8)
def extract_as_list(filename):
"""Opens file, appends each row (as dictionary) into list.
Parameters:
filename as string.
Returns:
output: list of each row in csv file as a dictionary.
"""
output = list()
csv_file = open(filename)
for row in csv.DictReader(csv_file):
output.append(row)
csv_file.close()
return output
def extract_as_dataframe(filename):
"""Reads in the data of the csv file into DataFrame.
Parameters:
filename: filename as string.
Returns:
data: DataFrame of csv file.
"""
data = pd.read_csv(filename, low_memory=False)
return data
def extract_data_for_months_by_year(data_list):
"""Extracts incidence per month for every year as a dictionary of dictionaries.
For example, data points for the months of year 1980 would look something like this:
{1980: {"January" : 18983, "February" : 23213, "March" : 31242 ... etc.}}
Parameters:
data_list: list of dictionaries, each item in list is a row of the CSV file.
Returns:
output_dict: dictionary mapping year to incidences per month of that year.
"""
output_dict = dict()
for row in data_list:
if row["Year"] in output_dict.keys():
year_dict = output_dict[row["Year"]]
if row["Month"] in year_dict.keys():
output_dict[row["Year"]][row["Month"]] += int(row["Incident"])
else:
output_dict[row["Year"]][row["Month"]] = int(row["Incident"])
else:
output_dict[row["Year"]] = dict()
assert (len(output_dict[row["Year"]].keys()) == 12), "Invalid number of months in data file."
return output_dict
def extract_data_by_categories(data_list, column_names, exclude_points):
""" Categorizes data points based on which column you want to look at.
Parameters:
data_list: list of dictionaries, each item in list is a row of the CSV file.
column_names: list of column names as strings that you want to investigate.
exclude_points: list of values as strings that are invalid data points such as "Unknown".
Returns:
output_dict: dictionary mapping each column to it's valid data points.
"""
output_dict = dict()
# Reading in rows of file, excluding points that are invalid or "placeholder" points
for row in data_list:
for name in column_names:
if row[name] not in exclude_points:
if name in output_dict.keys():
output_dict[name].append(row[name])
else:
output_dict[name] = [row[name]]
return output_dict
def find_mode_of_category(data_dict, column_name):
""" Given a specific column name, finds the most common data point (mode) and gives
its frequency.
Parameters:
data_dict: dictionary of selected column names as keys with their data points as values.
column_name: specific column name to find mode of.
Returns:
max_freq_datum: tuple consisting of: (data point that is the mode, frequency of that
data point).
"""
data = data_dict[column_name]
# Finding counts of each data point in column
number_dict = dict()
for datum in data:
if datum in number_dict.keys():
number_dict[datum] += 1
else:
number_dict[datum] = 1
# Converting counts into percentage of total data points (frequency)
num_of_points = sum(number_dict.values())
for val in number_dict:
number_dict[val] = float(number_dict[val]) / num_of_points
# Sorting data from max frequency to min frequency, first datum is max
max_to_min = sorted(number_dict.items(), key = itemgetter(1), reverse = True)
max_freq_datum = max_to_min[0]
return max_freq_datum
def find_max_of_all(data_dict, column_names):
""" Given multiple column names, finds the modes for each column and their frequencies.
Parameters:
data_dict: dictionary of selected column names as keys with their data points as values.
column_names: list of column names as strings that you want to investigate.
Returns:
output: list of tuples, each item in list as a return value from find_mode_of_category.
"""
output = list()
for name in column_names:
max_of_category = find_mode_of_category(data_dict, name)
output.append(max_of_category)
return output
def print_max_values(all_maxes, column_names):
"""Formats list returned from find_max_of_all for printing.
Parameters:
all_maxes: list returned from find_max_of_all, list of tuples as (mode datum, frequency).
column_names: list of column names as strings that you want to investigate.
Returns:
None, prints results from finding mode of the data.
"""
for val in range(len(column_names)):
category = column_names[val]
max_data = all_maxes[val][0]
percent = format(float(all_maxes[val][1]) * 100, '.2f')
print "Most affected %s: %s, %s" % (category, max_data, percent) + "%"
def extract_data_by_state(data_list, column_names, exclude_points, state_name):
"""Gets most modes for chosen columns for chosen state.
Parameters:
data_list: list of dictionaries, each item in list is a row of the CSV file.
column_names: list of column names as strings that you want to investigate.
exclude_points: list of values as strings that are invalid data points such as "Unknown".
state_name: state name as string that values are excluded to.
Returns:
modes_for_state: list of tuples as returned from find_max_of_all.
"""
state_list = list()
# Accumulate relevant rows of data_list for each state to be used to calculate mode
for row in data_list:
if row["State"] == state_name:
state_list.append(row)
accum_data = extract_data_by_categories(state_list, column_names, exclude_points)
modes_for_state = find_max_of_all(accum_data, column_names)
return modes_for_state
def print_state_data(data_list, column_names, exclude_points, states):
"""Formats calculated modes for printing.
Parameters:
data_list: list of dictionaries, each item in list is a row of the CSV file.
column_names: list of column names as strings that you want to investigate.
exclude_points: list of values as strings that are invalid data points such as "Unknown".
states: list of states to calculate modes for.
Returns:
None, prints values for each state chosen.
"""
for state in states:
state_modes = extract_data_by_state(data_list, column_names, exclude_points, state)
print "For the state of %s:" % (state)
print_max_values(state_modes, column_names)
print
def get_user_states(dict_states):
"""Asks user what states they would like to look at, makes sure that it is a valid state.
Parameters:
data_dict: dictionary of states mapping to their abbreviations.
Returns:
state_list: list of states user selected.
"""
print "Please capitalize the first letter of the state name."
state1 = str(raw_input('What is the first state you would like to look at? '))
state2 = str(raw_input('What is the second state you would like to look at? '))
print
state_list = [state1, state2]
for state in state_list:
assert (state in dict_states.keys()), "Please type a valid state name!"
return state_list
def spike_check_visual(year_data):
"""Orders months in chronological order and checks for spikes in total incidents per month.
Parameters:
year_data: dictionary of dictionaries, each year mapping to dictionary of months mapping
to total incidents.
Returns:
None, prints statements.
"""
year_set = set()
diff_dict = dict()
for input_y in range(1980, 2015):
input_year = str(input_y)
month_data = year_data[input_year].items()
x_val = [None] * 12
y_val = [None] * 12
# Order the months in chronological order so when graphing, x-axis does not display randomly
month = ["January", "February", "March", "April", "May", "June", "July", "August", "September",
"October", "November", "December"]
for items in month_data:
for i in range(12):
if items[0] == month[i]:
x_val[i] = items[0]
y_val[i] = items[1]
# Assigning each difference value with what to print if it becomes one of the three largest spikes.
for i in range(0, 11):
# A spike is defined as an increase in 150% or more over the time period of one month.
if (y_val[i] * 1.5) <= (y_val[i + 1]):
year_set.add(input_year)
diff = y_val[i + 1] - y_val[i]
diff_dict[diff] = "%s to %s %s" % (x_val[i], x_val[i + 1], input_year)
# Graphing only the years in which there was a spike, for visual confirmation.
if input_year in year_set:
graph_spike_year(x_val, y_val, input_year)
top_three_diff = sorted(diff_dict.keys(), reverse = True)[0:3]
print "Three largest spikes in total monthly incidents from 1980 to 2014:"
for diff in top_three_diff:
print diff_dict[diff]
print
def graph_spike_year(x_val, y_val, input_year):
"""Graphs each year which has a spike.
Parameters:
x_val: list of months.
y_val: list of total incidents for cooresponding month.
input_year: string that is the year which we are to graph.
Returns:
None, saves graph.
"""
pylab.figure(1)
x = range(12)
pylab.xticks(x, x_val)
pylab.plot(x, y_val, "g")
pylab.title("Number of Incidents per Month for " + input_year)
pylab.ylabel("Number of Incidents")
pylab.xlabel("Months")
pylab.savefig("yearly\incidents_" + str(input_year) + ".png")
pylab.clf()
def graph_affected_ages(age_clean_data):
"""Graphs plot showing number of incidents for different victim ages.
Parameters:
age_clean_data: cleaned DataFrame where age is not 998 or 99 (void values)
Returns:
None, saves graph.
"""
age_clean_data["Victim Age"].value_counts().sort_index(ascending = True).plot(kind = "bar",
color = "purple")
plt.title("Number of Incidents for Victim Ages")
plt.xlabel("Ages")
plt.ylabel("Number of Incidents")
plt.savefig("graphs\\victim_age.png")
plt.clf()
def graph_affected_sexes(data_frame):
"""Graphs plot showing number of incidents for different victim sexes.
Parameters:
data_frame: DataFrame of csv file.
Returns:
None, saves graph.
"""
data_frame["Victim Sex"].value_counts().plot(kind = 'bar')
plt.title("Number of Incidents for Victim Sexes")
plt.ylabel("Number of Incidents")
plt.savefig("graphs\\number_hom_sex.png")
plt.clf()
def graph_unsolved_cases_per_year(data_frame):
"""Graphs plot showing number of unsolved cases from 1980 to 2014.
Parameters:
data_frame: DataFrame of csv file.
Returns:
None, saves graph.
"""
unsolved = data_frame[data_frame["Crime Solved"] != "Yes"]
unsolved['Year'].value_counts().sort_index(ascending = True).plot(kind = 'line', color = "Red")
plt.title('Number of Unsolved Homicides: 1980 to 2014')
plt.savefig("graphs\unsolved_hom.png")
plt.clf()
def graph_weapons_handgun_over_time(data_frame):
"""Graphs number of cases with weapon documented as "handgun". Graphs handgun use over time.
Parameters:
data_frame: DataFrame of csv file.
Returns:
None, saves graph.
"""
ax2 = sns.countplot(x = "Year", hue = "Weapon", data = data_frame[data_frame["Weapon"] == "Handgun"],
palette = "colorblind")
ax2.legend(loc='upper right')
plt.title("Use of Handguns over Time")
plt.xlabel("Years")
plt.ylabel("Number of Homicides using Handguns")
plt.savefig("graphs\handgun_time.png")
plt.clf()
def find_incidents_for_states(data_list, dict_states):
"""Finds total number of incidents per state.
Parameters:
data_list: list of dictionaries, each item in list is a row of the CSV file.
dict_states: dictionary mapping each state name to it's abbreviation.
Returns:
final_dict: dictionary mapping each state to it's total number of incidents.
"""
final_dict = dict()
for row in data_list:
if dict_states[row["State"]] in final_dict.keys():
final_dict[dict_states[row["State"]]] += int(row["Incident"])
else:
final_dict[dict_states[row["State"]]] = int(row["Incident"])
return final_dict
def print_high_and_low_incidence(final_dict):
"""Prints state with highest total incidents and lowest.
Parameters:
final_dict: dictionary mapping each state to it's total incident count.
Returns:
None, prints.
"""
greatest_to_least = sorted(final_dict.items(), key = itemgetter(1), reverse = True)
print "States with highest and lowest total incidents:"
print "Highest: %s, Total Incidents: %d" %(greatest_to_least[0][0], greatest_to_least[0][1])
print "Lowest: %s, Total Incidents: %d" %(greatest_to_least[-1][0], greatest_to_least[-1][1])
print
def graph_crime_state(state_dict):
"""Graphs total incidents for each state.
Parameters:
state_dict: dictionary of states mapping to their abbreviation.
Returns:
None, saves graph.
"""
x_val = list()
y_val = list()
state_tup = state_dict.items()
for state in state_tup:
x_val.append(state[0])
y_val.append(state[1])
x = range(len(x_val))
pylab.xticks(x, x_val)
plt.xticks(rotation = 90)
pylab.bar(x, y_val, color='r')
pylab.title("Number of Incidents for all States")
pylab.ylabel("Number of Incidents")
pylab.xlabel("state")
pylab.savefig("graphs\incidents_state.png")
pylab.clf()
def main():
""" Will run main program when final_doc.py is run """
print "Welcome to Nanda and Vineeth's data analysis for all US homicides from 1980 - 2014"
print "Our data set has more than 600,000 data points, so please bear with us if the program is slow\n"
filename = "crime_data.csv"
column_names = ["Victim Sex", "Victim Age", "Victim Race", "Relationship"]
exclude_points = ["Unknown", "0", "998"]
dict_states = {
"Alaska": "AK", "Alabama": "AL", "Arkansas": "AR", "Arizona": "AZ", "California": "CA",
"Colorado": "CO","Connecticut": "CT", "District of Columbia": "DC", "Delaware": "DE",
"Florida": "FL", "Georgia": "GA","Hawaii": "HI", "Iowa": "IA", "Idaho": "ID", "Illinois": "IL",
"Indiana": "IN", "Kansas": "KS", "Kentucky": "KY", "Louisiana": "LA", "Massachusetts": "MA",
"Maryland": "MD", "Maine": "ME", "Michigan": "MI", "Minnesota": "MN", "Missouri": "MO",
"Mississippi": "MS", "Montana": "MT", "North Carolina": "NC", "North Dakota": "ND",
"Nebraska": "NE", "New Hampshire": "NH", "New Jersey": "NJ", "New Mexico": "NM","Nevada": "NV",
"New York": "NY", "Ohio": "OH", "Oklahoma": "OK", "Oregon": "OR", "Pennsylvania": "PA",
"Puerto Rico": "PR", "Rhodes Island": "RI", "South Carolina": "SC", "South Dakota": "SD",
"Tennessee": "TN", "Texas": "TX", "Utah": "UT", "Virginia": "VA", "Vermont": "VT",
"Washington": "WA", "Wisconsin": "WI", "West Virginia": "WV", "Wyoming": "WY"
}
# Extract data from CSV file
data_list = extract_as_list(filename)
# Find all data points for victim sex, race, age, and relationship
data_dict = extract_data_by_categories(data_list, column_names, exclude_points)
# Compare two state modes based on user input
answer = str(raw_input("Would you like to compare modes for two states? (yes/no) "))
assert (answer == "yes" or answer == "no"), "Please type a valid answer!"
print
if answer.lower() == "yes":
states = get_user_states(dict_states)
print_state_data(data_list, column_names, exclude_points, states)
else:
print "Please wait for the remaining computations!\n"
# Find most common datum for victim sex, race, age, and relationship
all_maxes = find_max_of_all(data_dict, column_names)
print "For all states:"
print_max_values(all_maxes, column_names)
print
print "Please wait, the program is graphing the results\n"
print "Plotting years with spikes...\n"
# Find incidence rate per month for every year
year_data = extract_data_for_months_by_year(data_list)
spike_check_visual(year_data)
# Extract data from CSV file as DataFrame
data_frame = extract_as_dataframe(filename)
print "Plotting total incidents for all states..."
# Graph all incidents for all states
incidents_dict = find_incidents_for_states(data_list, dict_states)
graph_crime_state(incidents_dict)
print "Done\n"
print_high_and_low_incidence(incidents_dict)
print "Plotting affected victim ages..."
# Graph affected victim ages
age_clean_data = data_frame[data_frame["Victim Age"] != 998]
age_clean_data = age_clean_data[age_clean_data["Victim Age"] != 99]
graph_affected_ages(age_clean_data)
print "Done\n"
print "Plotting use of handguns over time..."
# Graph use of handguns over time
graph_weapons_handgun_over_time(data_frame)
print "Done\n"
print "Plotting affected victim sex based on frequency..."
# Graph affected victim sex based on frequency
graph_affected_sexes(data_frame)
print "Done\n"
print "Plotting number of unsolved cases..."
# Graph number of unsolved cases from 1980 to 2014
graph_unsolved_cases_per_year(data_frame)
print "Done\n"
print "Check the folder 'yearly' and 'graphs' for the plots\n"
print "Program complete."
if __name__ == "__main__":
main()