-
Notifications
You must be signed in to change notification settings - Fork 2
/
main.py
376 lines (303 loc) · 15 KB
/
main.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
import wikipediaapi, requests, json, glob, string, nltk, contractions, re, os, random
from bs4 import BeautifulSoup
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
from warnings import simplefilter
from sklearn.feature_extraction.text import CountVectorizer
from wordcloud import WordCloud
from textwrap import wrap
def main():
# list all csv files only
generated_files = glob.glob('datasets/generated/*.{}'.format('csv'))
generated_json_files = glob.glob('datasets/generated/*.{}'.format('json'))
ihme_files = glob.glob('datasets/IHME-*.{}'.format('csv'))
countries = []
causes = []
# if global.csv does not exist, generate it
if not ('datasets/generated/global.json' in generated_json_files):
globaldf = generate_global_json(ihme_files)
else:
print('global.json already exists\nReading global.json')
globaldf = pd.read_json('datasets/generated/global.json')
if not ('datasets/generated/sample.json' in generated_json_files):
print('Generating sample.json')
sampledf = globaldf.sample(n=100)
sampledf.dropna(subset=['description'], inplace=True)
id_columns = [column for column in sampledf.columns if re.search("^[a-zA-Z]*_id$", column) is not None]
sampledf.drop(id_columns, axis = 'columns', inplace = True)
sampledf.to_json('datasets/generated/sample.json', orient="table")
else:
print('sample.json already exists')
dataset_description(globaldf)
""" commented out for quicker generation
characterize_dataset(globaldf)
# if cause_description.csv does not exist, generate it
if not ('datasets/generated/cause_description.csv' in generated_files):
causes = generate_causes_csv(globaldf)
else:
print('cause_description.csv already exists')
# if countries.csv does not exist, generate it
if not ('datasets/generated/countries.csv' in generated_files):
countries = generate_countries_csv(globaldf)
else:
print('countries.csv already exists')
if (len(countries) != 0) or (len(causes) != 0):
generate_individual_csvs(globaldf, causes, countries)
else:
print('Couldn\'t generate individual csvs')
"""
def generate_global_json(csv_files):
print('Generating global.json')
globaldf = pd.concat([pd.read_csv(file) for file in csv_files ], ignore_index=True)
numberdf = globaldf.loc[globaldf['metric_id'] == 1]
causes = numberdf["cause_name"].unique()
wiki = wikipediaapi.Wikipedia('Mozilla/5.0')
cause_description = pd.DataFrame(columns=['cause_name', 'description'])
for cause in causes:
page = wiki.page(cause)
# we have a "direct hit" for this cause
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
else:
# directly search wikipedia
url = 'https://en.wikipedia.org/w/index.php'
headers = {
'Accept' : '*/*',
'Accept-Language': 'en-US,en;q=0.5',
'User-Agent': 'Mozilla/5.0',
}
strip_cause = cause.lower().replace('other ', '')
parameters = {'search': strip_cause }
resp = requests.get(url, headers = headers, params = parameters)
# we may have a "direct hit" from removing "other"
if re.search("^https://en.wikipedia.org/wiki/" , resp.url):
page = wiki.page(strip_cause)
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
else:
# it wasn't a "direct hit", but we already have a valid search string on the first result
soup = BeautifulSoup(resp.text, 'html.parser')
search = soup.find_all('div', {"class":'mw-search-result-heading'})
search = re.sub(r'\([^)]*\)', '', search[0].text)
page = wiki.page(search)
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
countries = numberdf["location_name"].unique()
countries_cca3 = pd.DataFrame(columns=['location_name', 'cca3'])
error_countries = {
"Venezuela (Bolivarian Republic of)": "VEN",
"Taiwan (Province of China)": "TWN",
"Iran (Islamic Republic of)": "IRN",
"Iraq (Republic of)": "IRQ",
"Ireland (Republic of)": "IRL",
"Bolivia (Plurinational State of)": "BOL",
"Micronesia (Federated States of)": "FSM",
}
for country in countries:
cca3_r = requests.get(url = f'https://restcountries.com/v3.1/name/{country}')
if country == "China":
cca3 = "CHN"
countries_cca3 = countries_cca3._append({'location_name': country, 'cca3': cca3}, ignore_index=True)
elif cca3_r.status_code == 200:
cca3_data = cca3_r.json()
cca3 = cca3_data[0]['cca3']
countries_cca3 = countries_cca3._append({'location_name': country, 'cca3': cca3}, ignore_index=True)
else:
if country in error_countries:
cca3 = error_countries[country]
countries_cca3 = countries_cca3._append({'location_name': country, 'cca3': error_countries[country]}, ignore_index=True)
else:
print("Couldn't find cca3 for", country)
continue
# gjson_r = requests.get(url = f'http://inmagik.github.io/world-countries/countries/{cca3}.geojson')
# if gjson_r.status_code == 200:
# print("Found GEOJSON for", cca3, country)
# gjson = gjson_r.json()
# else:
# print("Couldn't find GEOJSON for", cca3, country)
# continue
# try:
# with open(f'gjsons/{cca3}.json', 'x', encoding='utf-8') as f:
# json.dump(gjson, f, ensure_ascii=False, indent=4)
# except FileExistsError:
# print("GEOJSON already exists", cca3, country)
numberdf['cca3'] = numberdf['location_name'].map(countries_cca3.set_index('location_name')['cca3'])
numberdf['description'] = numberdf['cause_name'].map(cause_description.set_index('cause_name')['description'])
sampledf = numberdf.sample(n=10000)
id_columns = [column for column in numberdf.columns if re.search("^[a-zA-Z]*_id$", column) is not None]
sampledf.drop(id_columns, axis = 'columns', inplace = True)
numberdf.to_json('datasets/generated/global.json', orient='records')
sampledf.to_json('datasets/generated/sample.json', orient='records')
dataset_description(numberdf)
print('global.json generated')
return numberdf
def generate_causes_csv(globaldf):
print("Generating cause_description.csv")
causes = globaldf["cause_name"].unique()
descriptions = []
info_causes = []
wiki = wikipediaapi.Wikipedia('Mozilla/5.0')
cause_description = pd.DataFrame(columns=['cause_name', 'description']);
for cause in causes:
page = wiki.page(cause)
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
info_causes.append(cause)
description = fix_description(description)
generate_graphs(description, cause)
descriptions += description
else:
url = 'https://en.wikipedia.org/w/index.php'
headers = {
'Accept' : '*/*',
'Accept-Language': 'en-US,en;q=0.5',
'User-Agent': 'Mozilla/5.0',
}
strip_cause = cause.lower().replace('other ', '')
parameters = {'search': strip_cause }
resp = requests.get(url, headers = headers, params = parameters)
if re.search("^https://en.wikipedia.org/wiki/" , resp.url):
page = wiki.page(strip_cause)
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
info_causes.append(cause)
description = fix_description(description)
generate_graphs(description, cause)
descriptions += description
else:
soup = BeautifulSoup(resp.text, 'html.parser')
search = soup.find_all('div', {"class":'mw-search-result-heading'})
search = re.sub(r'\([^)]*\)', '', search[0].text)
page = wiki.page(search)
if page.exists():
description = page.summary
cause_description = cause_description._append({'cause_name': cause, 'description': description}, ignore_index=True)
info_causes.append(cause)
description = fix_description(description)
generate_graphs(description, cause)
descriptions += description
generate_graphs(descriptions, "global")
cause_description.to_csv('datasets/generated/cause_description.csv', index=False)
print("cause_description.csv generated")
return info_causes
def generate_countries_csv(globaldf):
print("Generating countries.csv")
countries = globaldf["location_name"].unique()
info_countries = []
all_countrydf = pd.DataFrame(columns=['country', 'code'])
for country in countries:
cca3_r = requests.get(url = f'https://restcountries.com/v3.1/name/{country}')
if cca3_r.status_code == 200:
cca3_data = cca3_r.json()
cca3 = cca3_data[0]['cca3']
else:
continue
gjson_r = requests.get(url = f'http://inmagik.github.io/world-countries/countries/{cca3}.geojson')
if gjson_r.status_code == 200:
gjson = gjson_r.json()
else:
continue
try:
with open(f'gjsons/{cca3}.geojson', 'x', encoding='utf-8') as f:
json.dump(gjson, f, ensure_ascii=False, indent=4)
except FileExistsError:
print("GEOJSON already exists")
all_countrydf = all_countrydf._append({'country': country, 'code': cca3}, ignore_index=True)
info_countries.append((country, cca3))
all_countrydf.to_csv('datasets/generated/countries.csv', index=False)
print("countries.csv generated")
return info_countries
def generate_individual_csvs(globaldf, causes, countries):
for (country, cca3) in countries:
countrydf = globaldf.loc[globaldf['location_name'] == country]
countrydf.to_csv(f"datasets/generated/countries/{cca3}.csv", index=False)
for year in globaldf['year'].unique():
yeardf = globaldf.loc[globaldf['year'] == year]
yeardf.to_csv(f"datasets/generated/year/{year}.csv", index=False)
for cause in causes:
causedf = globaldf.loc[globaldf['cause_name'] == cause]
causedf.to_csv(f"datasets/generated/causes/{cause.replace('/', '_')}.csv", index=False)
def fix_description(description):
cleaned_string = contractions.fix(description).lower().translate(str.maketrans('', '', string.punctuation)).replace('\n', '')
word_list = re.findall('\w+', cleaned_string)
stopwords = nltk.corpus.stopwords.words('english')
words = []
for word in word_list:
if word not in stopwords:
words.append(word)
return words
def generate_graphs(description, cause):
fdist = nltk.FreqDist(word for word in description)
if cause == "global":
df_fdist = pd.DataFrame.from_dict(fdist, orient='index')
df_fdist.columns = ['Frequency']
df_fdist = df_fdist.sort_values(by=['Frequency'], ascending=False)
df_fdist.to_csv('datasets/generated/global_word_frequency.csv', index=True)
generate_bars(fdist, cause)
generate_wordcloud(fdist, cause)
def color_func(word, font_size, position, orientation, random_state=None,
**kwargs):
return f"hsl({random.randint(0, 20)}, 100%, {random.randint(60, 100)}%)"
def generate_wordcloud(fdist, cause):
wc = WordCloud(width=800, height=400, max_words=20, stopwords=nltk.corpus.stopwords.words('english'), color_func=color_func).generate_from_frequencies(fdist)
plt.figure(figsize=(10, 10))
plt.imshow(wc, interpolation='bilinear')
plt.axis('off')
plt.savefig(f"wordclouds/{cause.replace('/', '_')}.png", bbox_inches='tight', pad_inches=0)
plt.close()
def generate_bars(fdist, cause):
series = pd.Series(dict(fdist.most_common(20)))
fig, ax = plt.subplots(figsize=(10,10))
plot = sns.barplot(x=series.index, y=series.values, ax=ax, palette="flare", hue=series.index)
plt.xticks(rotation=30);
plot.set(xlabel = "Term", ylabel = "Frequency", title = f"Most common terms for {cause.lower()}")
plot.figure.savefig(f"graphs/{cause.replace('/', '_')}.png", bbox_inches='tight', pad_inches=0)
plt.close()
def dataset_description(df):
f = open("datasets/generated/dataset_description.txt", "w")
f.write(str(df.describe()))
f.write("\n\n")
for column in df.columns:
f.write(f"{column}:\n")
f.write(str(df[column].describe()))
f.write("\n")
f.write(f"# unique entries {column}: {str(len(df[column].unique()))}")
f.write("\n\n")
f.write("\n\n")
f.write(f"{df['cause_name'].unique()}")
f.close()
idx = df.groupby('cause_name')['val'].idxmax()
cause_maxvals = df.loc[idx]
cause_maxvals.to_csv('datasets/generated/cause_maxvals.csv', index=False)
idx = df.groupby('location_name')['val'].idxmax()
country_maxvals = df.loc[idx]
country_maxvals.to_csv('datasets/generated/country_maxvals.csv', index=False)
idx = df.groupby('year')['val'].idxmax()
year_maxvals = df.loc[idx]
year_maxvals.to_csv('datasets/generated/year_maxvals.csv', index=False)
year_maxvals['val'] = year_maxvals['val'].astype(int)
plot = sns.barplot(x='year', y='val', data=year_maxvals)
plt.xticks(rotation=30);
plot.set(xlabel = "Year", ylabel = "Deaths", title = f"Number of deaths by year")
plot.figure.savefig(f"datasets/generated/deaths_year.png", bbox_inches='tight', pad_inches=0)
plt.close()
def characterize_dataset(df):
top_mortality_ranking(df)
return
def top_mortality_ranking(df):
df = df.loc[df['year'] == 2019]
pd.options.display.float_format = '{:.2f}'.format
df = df[['cause_name', 'val', 'upper', 'lower']].groupby(['cause_name']).sum()
ranking = df.nlargest(10, 'val')
ranking.to_csv(f'datasets/generated/top_10_deadliest_diseases.csv')
if __name__ == '__main__':
nltk.download("stopwords")
nltk.download("punkt")
simplefilter(action = 'ignore', category = FutureWarning)
main()