forked from Covid-19-Response-Greece/covid19-region-forecast
-
Notifications
You must be signed in to change notification settings - Fork 0
/
app_functions.py
178 lines (155 loc) · 8.04 KB
/
app_functions.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
"""
@author: Elena Stamatelou
"""
import pandas as pd
# pip install fbprophet
from fbprophet import Prophet
import logging
logging.getLogger('fbprophet').setLevel(logging.WARNING) # remove warning messages in prophet
#import subprocess
#from pathlib import Path
import datetime as dt
# Define functions
#!pip install PyGithub
# generate GitHub API token
def read_from_github():
# Input: no
# Outputs: the main datasets ready to use
# Function: reads the files from GitHub
# Online Github path
from github.MainClass import Github
token = 'f1d4c930800cb3e782c1b61b86101c97deefc4c7'
# connected to GitHub using my credentials
g = Github(token)
# load the repo
repo = g.get_repo("CSSEGISandData/COVID-19")
file_list = repo.get_contents("csse_covid_19_data/csse_covid_19_time_series")
github_dir_path = 'https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_time_series/'
file_path_confirmed_US = github_dir_path + str(file_list[-5]).split('/')[-1].split(".")[0]+ '.csv'
confirmed_US = pd.read_csv(file_path_confirmed_US, error_bad_lines=False)
confirmed_US = confirmed_US.drop(['UID', 'iso2', 'iso3', 'code3', 'FIPS', 'Admin2', 'Combined_Key'], axis=1)
confirmed_US=confirmed_US.rename(columns = {'Province_State':'Province/State',
'Country_Region':'Country/Region',
'Long_':'Long'})
confirmed_info = confirmed_US.groupby(["Province/State"])['Lat','Long','Country/Region'].agg({
'Lat' :lambda x: x.mean(),
'Long' :lambda x: x.mean(),
'Country/Region':lambda x: x.unique()
})
cases = confirmed_US.drop(columns=['Country/Region', 'Lat', 'Long']).set_index('Province/State')
cases = cases.groupby(["Province/State"]).sum()
confirmed_US = confirmed_info.merge(cases, left_index=True, right_index=True).reset_index()
confirmed_US = confirmed_US[confirmed_US['Long'].isna() == False]
confirmed_US = confirmed_US[confirmed_US['Lat'].isna() == False]
confirmed_US = confirmed_US[confirmed_US['Long'] != 0]
confirmed_US = confirmed_US[confirmed_US['Lat'] != 0]
file_path_confirmed = github_dir_path + str(file_list[-4]).split('/')[-1].split(".")[0]+ '.csv'
confirmed = pd.read_csv(file_path_confirmed, error_bad_lines=False)
confirmed = pd.concat([confirmed_US,confirmed])
file_path_died_US = github_dir_path + str(file_list[-3]).split('/')[-1].split(".")[0]+ '.csv'
died_US = pd.read_csv(file_path_died_US, error_bad_lines=False)
died_US = died_US.drop(['UID', 'iso2', 'iso3', 'code3', 'FIPS', 'Admin2', 'Combined_Key', 'Population'], axis=1)
died_US=died_US.rename(columns = {'Province_State':'Province/State',
'Country_Region':'Country/Region',
'Long_':'Long'})
died_info = died_US.groupby(["Province/State"])['Lat','Long','Country/Region'].agg({
'Lat' :lambda x: x.mean(),
'Long' :lambda x: x.mean(),
'Country/Region':lambda x: x.unique()
})
deaths = died_US.drop(columns=['Country/Region', 'Lat', 'Long']).set_index('Province/State')
deaths = deaths.groupby(["Province/State"]).sum()
died_US = died_info.merge(deaths, left_index=True, right_index=True).reset_index()
died_US = died_US[died_US['Long'].isna() == False]
died_US = died_US[died_US['Lat'].isna() == False]
died_US = died_US[died_US['Long'] != 0]
died_US = died_US[died_US['Lat'] != 0]
#
file_path_died = github_dir_path + str(file_list[-2]).split('/')[-1].split(".")[0]+ '.csv'
died = pd.read_csv(file_path_died, error_bad_lines=False)
died = pd.concat([died_US,died])
file_path_recovered = github_dir_path + str(file_list[-1]).split('/')[-1].split(".")[0]+ '.csv'
recovered = pd.read_csv(file_path_recovered, error_bad_lines=False)
# preprocessing to add total column
confirmed = preprocessing(confirmed)
died = preprocessing(died)
recovered = preprocessing(recovered)
# demographics
population_statistics = pd.read_csv("population_statistics.csv")
return confirmed, died, recovered, population_statistics
def save(confirmed, died, recovered):
# save the files
a = dt.datetime.now()
confirmed.to_csv("time_series_19-covid-Confirmed.csv",index=False)
died.to_csv("time_series_19-covid-Deaths.csv",index=False)
recovered.to_csv("time_series_19-covid-Recovered.csv",index=False)
print("save_main_files",dt.datetime.now()-a)
return
def read_files():
# Input: no
# Outputs: the main datasets ready to use
# Function: reads the files
## Confirmed ##
confirmed = pd.read_csv("time_series_19-covid-confirmed.csv")
confirmed = preprocessing(confirmed)
## Died ##
died = pd.read_csv("time_series_19-covid-Deaths.csv")
died = preprocessing(died)
## Recovered ##
recovered = pd.read_csv("time_series_19-covid-Recovered.csv")
recovered = preprocessing(recovered)
population_statistics = pd.read_csv("population_statistics.csv")
return confirmed, died, recovered, population_statistics
def preprocessing(df):
# Inputs: Inputs datasets of confirmed, died,and recovered cases
# Ouputs: Preprocessed datasets
# Function: Creates an extra column with the total number of cases per country/row
number = df.drop(['Province/State', 'Country/Region', 'Lat' ,'Long'], axis=1)
number.loc[:,'total'] = number.sum(axis=1)
df = pd.concat([df,number.total],axis = 1)
return df
def selected_countries_df(df, selected_countries):
# Inputs: All the dataset and the selected countries (an area is selected in the map)
# Outputs:The data of the selected countries
# Function: Filters the dataset
a = dt.datetime.now()
selected_countries_data_all = pd.DataFrame()
for i in range(len(selected_countries)):
selected_data = df[df['Country/Region'] == selected_countries[i]]
selected_countries_data_all = pd.concat([selected_countries_data_all, selected_data])
selected_countries_data_all = selected_countries_data_all.drop(['Province/State', 'Country/Region', 'Lat' ,'Long', 'total'], axis=1).reset_index(drop=True)
selected_countries_data_all.loc['total',:]= selected_countries_data_all.sum(axis=0)
selected_countries_data_all = selected_countries_data_all.T
selected_countries_data_all = selected_countries_data_all.reset_index()
selected_countries_data_all['index'] = pd.to_datetime(selected_countries_data_all['index'])
print("Filters the dataset",dt.datetime.now()-a)
return selected_countries_data_all
def predictions(df):
# Input: time series dataframe with two columns (date, values)
# Output: predictions
# Function: applies prophet forecast method
a = dt.datetime.now()
predictions = df[['index','total']]
predictions.columns = ['ds', 'y']
# predictions = predictions.iloc[:-1,:]
# create the model
# m = Prophet(yearly_seasonality=True)
# add monthly seasonality
# m.add_seasonality(name="monthly",period=30.5,fourier_order=5)
m = Prophet(changepoint_prior_scale=0.2, changepoint_range=0.98,uncertainty_samples=0, yearly_seasonality=False, seasonality_mode='additive')
m.add_seasonality(name="monthly",period=30.5,fourier_order=5)
# predict
m.fit(predictions)
future = m.make_future_dataframe(periods=5)
fcst = m.predict(future)
fcst['yhat'] = fcst['yhat'].astype(int)
print("Predict",dt.datetime.now()-a)
return fcst
def difference_prediction_actual(df, fcst):
actual = df[['index','total']]
actual.columns = ['ds', 'y']
forecasted = fcst[['ds','yhat']]
actual_and_forecasted = pd.merge(left=actual, right=forecasted, left_on='ds', right_on='ds')
actual_and_forecasted['difference'] = abs((actual_and_forecasted['y'] - actual_and_forecasted['yhat'])/actual_and_forecasted['y'])
mape = actual_and_forecasted['difference'].sum() / len(actual_and_forecasted)
return mape