-
Notifications
You must be signed in to change notification settings - Fork 0
/
general_data_parser.py
299 lines (216 loc) · 11.2 KB
/
general_data_parser.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
import pandas as pd
from pprint import pprint
pd.set_option('display.max_columns', 30)
pd.set_option('display.max_rows', 190)
'''
#Ön hazırlık kodları (corona statsı, HDI ve healthexpenseperGDP datalarını birleştirmek için)
hdi = "datasets/Human development index (HDI).csv"
coronastats = "datasets/latest_covid_data.csv"
HealthexpenseperGDP= "datasets/HealthexpenseperGDP.csv"
hdi_df = pd.read_csv(hdi, header=1)
hdi_df.set_index('Country', inplace=True)
corona_df = pd.read_csv(coronastats)
corona_df.rename(columns={'location': 'Country'}, inplace=True)
corona_df.set_index('Country', inplace=True)
#print(corona_df)
HpG_df = pd.read_csv(HealthexpenseperGDP)
HpG_filter = (HpG_df['TIME'] >= 2018) #& ((HpG_df['LOCATION'] == 'TUR') | (HpG_df['LOCATION'] == 'ITA') | (HpG_df['LOCATION'] == 'USA'))
HpG_df = HpG_df.loc[HpG_filter, ['LOCATION', 'TIME', 'Value']]
HpG_df.rename(columns={'Value':'HealthExpenseperGDP', 'LOCATION':'iso_code'}, inplace=True)
HpG_df.set_index('iso_code', inplace=True)
#print(HpG_df)
data= corona_df.join(hdi_df['2018'], how='outer')
data.rename(columns = {'2018': 'HDI'}, inplace= True)
data.reset_index(inplace=True)
data.set_index('iso_code', inplace=True)
data= data.join(HpG_df['HealthExpenseperGDP'], how='outer')
#print(data)
data_for_stats = data
data_for_stats.reset_index(inplace=True)
data_for_stats.to_csv("all_dataappended.txt", sep=",", index=False, header=True)
'''
#DATA preprocessing - Handling non-typical values
corona_file = r'C:\Users\BERK\PycharmProjects\CORONA_PROJECT\datasets\all_corona_data.txt'
corona_dataframe = pd.read_table(corona_file)
def str_float_convert(str_type, index, column):
'''
data içindeki verileri düzenler (specific olarak corona datası)
:param str_type: dataframe column (datamalipulator'dan alır)
:param index: dataframe index (data_manipulator'dan alır)
:param column: ref value alabilmek icin ihtiyac duyar
:return: float ve rounded values
'''
try:
f_type = round(float(str_type), 3)
if f_type:
corona_dataframe.loc[index, column] = f_type
return f_type
except ValueError:
s_list = str_type.split('.')
if 'aged' in column or 'handwash' in column or 'cvd' in column or 'Expense' in column:
s_str = s_list[0] + '.' + ''.join(s_list[1:])
f_str = round(float(s_str), 3)
return f_str
if 'gdp' in column:
s_str = ''.join(s_list[0:2]) + '.' + ''.join(s_list[2])
f_str = round(float(s_str), 3)
return f_str
if index == 0: return str_type
else:
ref_value = corona_dataframe.loc[index - 1, column] #bazı column'larda uygunsuz değerleri değiştirebilmek için bir referans değerine ihtiyaç duydum.
ref_list = str(ref_value).split('.')
if 'density' in column:
ref_value = 237.016
ref_list = str(ref_value).split('.')
s_str = "".join(s_list)
s_str = s_str[0:len(ref_list[0])] + '.' + s_str[len(ref_list[0]):]
f_str = round(float(s_str), 3)
return f_str
if len(s_list) != len(ref_list):
s_str = "".join(s_list)
s_str= s_str[0:len(ref_list[0])] + '.' + s_str[len(ref_list[0]):]
f_str = round(float(s_str), 3)
corona_dataframe.loc[index, column] = f_str
return f_str
return '00000'
def data_manipulator(column_name, dataframe):
'''
corona datatsında bazı veriler new_death'i 0 olmasına rağmen 1-2 hafta öncesinden girdilere sahip.
Bunları elimine etmek ve tüm ülkeleri ilk case- ilk date olarak sabitlemek
:param tsv_file: birleştirilmiş corona datası (tsv) formatında
:return: pandas dataframe
'''
df = dataframe
df = df[df['total_cases'] != 0]
df = df[df['total_deaths'] != 0]
c_name = column_name
df[c_name] = [i for i in map(str_float_convert, df[c_name], df.index, [c_name for _ in range(len(df[c_name]))])]
return df
'''
manipulated_df = data_manipulator('total_cases_per_million', corona_dataframe)
manipulated_df = data_manipulator('total_deaths_per_million', manipulated_df)
manipulated_df = data_manipulator('total_cases_per_million', manipulated_df)
manipulated_df = data_manipulator('total_tests_per_thousand', manipulated_df)
manipulated_df = data_manipulator('new_deaths_per_million', manipulated_df)
manipulated_df = data_manipulator('new_cases_per_million', manipulated_df)
manipulated_df = data_manipulator('new_tests_per_thousand', manipulated_df)
manipulated_df= data_manipulator('aged_70_older', manipulated_df)
manipulated_df= data_manipulator('aged_65_older', manipulated_df)
manipulated_df= data_manipulator('gdp_per_capita', manipulated_df)
manipulated_df= data_manipulator('population_density', manipulated_df)
manipulated_df= data_manipulator('handwashing_facilities', manipulated_df)
manipulated_df= data_manipulator('cvd_death_rate', manipulated_df)
manipulated_df= data_manipulator('HealthExpenseperGDP', manipulated_df)
'''
#manipulated_df.to_csv('All_manipulated.txt', index=False, sep='\t')
# Further manipulation of the data and filling mising values
import math
'''
df_file = 'All_manipulated_cleanedUp.txt'
df = pd.read_table(df_file)
country_prot_number = {'China': 574, 'Australia': 1431, 'United States': 6390, 'United Kingdom': 15006,
'Netherlands': 598, 'Czech Republic': 30, 'Taiwan': 104, 'Luxembourg': 257,
'India': 363, 'South Korea': 33, 'Iran': 8, 'Nigeria': 1, 'Mexico': 17, 'Italy': 79,
'Iceland': 601, 'Thailand': 125, 'Turkey': 63, 'Finland': 40, 'Portugal': 100, 'France': 391,
'Democratic Republic of the Congo': 126, 'Japan': 130, 'Germany': 201, 'New Zealand': 8,
'Sweden': 163, 'Belgium': 509, 'Switzerland': 76, 'Spain': 497, 'Jordan': 28, 'Russia': 207,
'Brazil': 95, 'Sri Lanka': 4, 'Malaysia': 16, 'Colombia': 86, 'Qatar': 16, 'Lithuania': 3,
'Hungary': 32, 'Kuwait': 8, 'Georgia': 15, 'Argentina': 29, 'Panama': 1, 'Poland': 27,
'Croatia': 7, 'Vietnam': 28, 'Saudi Arabia': 130, 'Canada': 208, 'Philippines': 13,
'Slovakia': 4, 'Egypt': 2, 'Uruguay': 11, 'Peru': 2, 'Ireland': 18, 'Chile': 144, 'Latvia': 25,
'Singapore': 159, 'Gambia': 3, 'Slovenia': 5, 'Denmark': 627, 'Korea': 3, 'South Africa': 20,
'Estonia': 5, 'Algeria': 3, 'Greece': 99, 'Senegal': 23, 'Austria': 250, 'Belarus': 2,
'Norway': 48, 'Pakistan': 2, 'Ecuador': 4, 'Cambodia': 1, 'Nepal': 1, 'Ghana': 15,
'Israel': 222, 'Costa Rica': 6, 'Democratic Republic of the Congo ': 7,
'Kazakhstan': 4, 'United Arab Emirates': 25, 'Indonesia': 9, 'Brunei': 5,
'Serbia': 4, 'Myanmar': 1, 'Bangladesh': 20, 'Guam': 3, 'Romania': 3, 'Tunisia': 3,
'Lebanon': 10, 'Jamaica': 8, 'Uganda': 20}
country_names = [c for c in country_prot_number.keys()]
country_filter = df['Country'].isin(country_names)
df['log_population'] = [i for i in map(math.log, df['population'])]
new_df = df[country_filter]
new_df['new_deaths'].fillna(0, inplace=True)
new_df['new_cases'].fillna(0, inplace=True)
new_df['new_tests'].fillna(0, inplace=True)
#print(new_df['new_cases_per_million'].describe())
new_df['new_cases_per_million'].fillna((new_df['new_cases']/new_df['population'])*1000000, inplace=True)
new_df['new_deaths_per_million'].fillna((new_df['new_deaths']/new_df['population'])*1000000, inplace=True)
new_df['total_cases_per_million'].fillna((new_df['total_cases']/new_df['population'])*1000000, inplace=True)
new_df['new_tests_per_thousand'].fillna((new_df['new_tests']/new_df['population'])*1000, inplace=True)
new_df['total_tests_per_thousand'].fillna((new_df['total_tests']/new_df['population'])*1000, inplace=True)
#print(new_df.isnull().sum())
'''
def rounder_func(df, column_name):
floating_lst = [float(f) for f in df[column_name]]
rounded = [round(x, 3) for x in floating_lst]
df[column_name] = [r for r in rounded]
return df
'''
rounded_new_df = rounder_func(new_df, 'new_cases_per_million')
rounded_new_df = rounder_func(rounded_new_df, 'new_deaths_per_million')
rounded_new_df = rounder_func(rounded_new_df, 'total_cases_per_million')
rounded_new_df = rounder_func(rounded_new_df, 'total_deaths_per_million')
rounded_new_df = rounder_func(rounded_new_df, 'population_density')
rounded_new_df = rounder_func(rounded_new_df, 'aged_70_older')
rounded_new_df = rounder_func(rounded_new_df, 'aged_65_older')
rounded_new_df = rounder_func(rounded_new_df, 'log_population')
#rounded_new_df = rounder_func(new_df, 'new_tests_per_thousand')
#rounded_new_df = rounder_func(new_df, 'total_tests_per_thousand')
rounded_new_df.drop(
columns=['new_tests_per_thousand', 'total_tests_per_thousand', 'total_tests', 'new_tests_smoothed',
'new_tests_smoothed_per_thousand', 'tests_units', 'new_tests'], inplace=True
)
#print(new_df.describe())
#print(new_df.isnull().sum())
#rounded_new_df.to_csv('Filled_cleaned_coronadf.txt', index=False, sep='\t')
'''
#Adding smokers data
smokers_file = pd.read_csv(r'C:\Users\BERK\PycharmProjects\CORONA_PROJECT\datasets\smokers.csv')
smoker_df = smokers_file.drop(columns=['maleSmokingRate','femaleSmokingRate', 'pop2020'])
#cigar_filter = smoker_df['Country'] == ()
smoker_df.rename(columns={'name':'Country', 'totalSmokingRate':'total_smokers'}, inplace=True)
#print(smoker_df)
#rounded_new_df = rounded_new_df.merge(smoker_df, on='Country', how='outer')
#df=df.merge(smoker_df, on='Country', how='outer')
#df.to_csv('smoker_added_coronadfall.txt', index=False, sep='\t')
#print(rounded_new_df)
#rounded_new_df['iso_code'].dropna(inplace=True)
'''
def total_smoker_fill(df):
sex_ratios = { #country: males/females
'Algeria': 102.1/100, 'Czech Rebuplic': 96.984/100, 'Gambia':98.393/100, 'Guam': 101.79/100,
'Kuwait': 157.866/100, 'New Zealand': 96.55/100, 'Peru': 98.707/100, 'Qatar': 302.426/100,
'Taiwan': 98.764/100, 'Tunisia': 98.37/100, 'United Arab Emirates': 223.845/100
}
data = df
for country in sex_ratios:
filter = data['Country'] == country
data.loc[filter, 'total_smokers'] = (data.loc[filter, 'female_smokers']*sex_ratios[country] + data.loc[filter,'male_smokers'])/(sex_ratios[country]+1)
filled_df = data
return filled_df
cigarated_df = total_smoker_fill(rounded_new_df)
cigarated_df = rounder_func(cigarated_df, 'total_smokers')
cigarated_df['iso_code'].dropna(inplace=True)
print(cigarated_df['total_smokers'].describe())
print(cigarated_df['total_smokers'].isna().sum())
print(cigarated_df)
cigarated_df.to_csv('smoker_added_coronadf.txt', index=False, sep='\t')
'''
#Protein datasını preprocess edebilme
protein_file = r"C:\Users\BERK\PycharmProjects\CORONA_PROJECT\datasets\ALL11111dataframe.csv"
def percent_float_converter(file):
with open(file, 'r') as f:
with open('all_proteins.txt', 'w') as at:
for line in f:
new_line = line.strip().replace('%', '')
at.write(new_line+'\n')
#percent_float_converter(protein_file)
protein_dataframe = pd.read_csv('all_proteins.txt', header=0, sep='\t')
protein_df = protein_dataframe.drop(columns=['id'])
protein_df.rename(columns={'country':'Country'}, inplace=True)
print(protein_df.describe())
#print(protein_df['identity for NSP1'].describe())
print(protein_df.isna().sum())
#fil_tr = protein_df['Country'] == 'Turkey'
#protein_tr = protein_df[fil_tr]
#print(protein_tr)