Merge branch 'ServiceChargePrediction' into 'master'

maintenance date changes

See merge request !34

Email_Generation.py

+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.linear_model import LogisticRegression
+import warnings
+
+warnings.filterwarnings("ignore")
+
+df = pd.read_csv('Data/MaintainData.csv')
+#df = df.groupby('mailID')
+
+#df = pd.DataFrame(df['Customer'].unique())
+
+df['a'] = pd.to_datetime(df['Date']).diff()
+df['a'] = df['a'].astype(str).str.replace("days", "")
+df['a'] = df['a'].astype(str).str.replace("NaT", "30")
+df = df[(df["mailID"] == 'nbachalaka@gmail.com')]
+X = df[['TakenTime', 'Price', 'MeterReading']]
+y = df['a']
+
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)
+
+logreg = LogisticRegression()
+logreg.fit(X_train, y_train)
+Accuracy = logreg.score(X_test, y_test)
+
+MeterReading = df['MeterReading'].mean()
+TakenTime = df['TakenTime'].mean()
+Price = df['Price'].mean()
+
+print(Price)
+new_input = [[TakenTime, Price, MeterReading]]
+pred = logreg.predict(X)
+# print(pred)
+
+# Assign predicted Dates to datafram
+se = pd.Series(pred)
+df['new_col'] = se.values
+
+
+df['Date'] = pd.to_datetime(df['Date'])
+df['ComesON'] = df['Date'] + pd.to_timedelta(df['new_col'].astype(int), unit='D')
+print(df)
+
+# Get Today Date
+from datetime import date
+today = date.today()
+print(today)
+# dd/mm/YY
+d1 = today.strftime("%d/%m/%Y")
+print(d1)
+# 3 days before predicted date
+enddate = pd.to_datetime(d1) - pd.DateOffset(days=3)
+
+print(enddate)
+# print(enddate.strftime("%d/%m/%Y"))
+
+
+
+finalDate = enddate.strftime("%d/%m/%Y")
+# df = df[(df["ComesON"] == finalDate)]
+print(df)
+PredictedMails = df["mailID"].tolist()
+#print(PredictedMails)
+
+
+
+# mail server
+import smtplib
+gmail_user = 'chalakasliit@gmail.com'
+gmail_password = 'SLIIT2022%'
+
+sent_from = gmail_user
+to = PredictedMails # customer mails
+subject = 'Reminder about Your Vehicle Service'
+body = 'Still 3 Days for your Vehicle service!'
+
+email_text = """\
+From: %s
+To: %s
+Subject: %s
+
+%s
+""" % (sent_from, ", ".join(to), subject, body)
+
+try:
+    server = smtplib.SMTP_SSL('smtp.gmail.com', 465)
+    server.ehlo()
+    server.login(gmail_user, gmail_password)
+    server.sendmail(sent_from, to, email_text)
+    server.close()
+
+    print("Email sent")
+except:
+    print("No services available...")
+
+# mail server
+
+
+

User/views.py

@@ -838,7 +838,7 @@ def NextMaintain(request):
         myForm = MyFormServices(request.POST)
         #myForm3 = MyForm3(request.POST or None)
 
-# price  prediction with logitic
+#Next Service Date Prediction with logistic Regression
         if myForm.is_valid():
 
             customerid = myForm.cleaned_data['customerid']
@@ -862,8 +862,6 @@ def NextMaintain(request):
                 dataset = df.head()
                 # print(dataset)
                 # do same but attach it to the dataframe
-                # df['Different'] = df.apply(lambda row: 45, axis=1)
-
                 df['a'] = pd.to_datetime(df['Date']).diff()
                 df['a'] = df['a'].astype(str).str.replace("days", "")
                 df['a'] = df['a'].astype(str).str.replace("NaT", "30")
@@ -887,18 +885,16 @@ def NextMaintain(request):
                 pred = int(pred)
                 print("--------------Customer will come after days---------------------")
                 # print(pred)
-                startdate = "7/7/2021"
+                startdate = "10/31/2021"
                 enddate = pd.to_datetime(startdate) + pd.DateOffset(days=pred)
                 print(enddate)
 
 
                 df = df[:1]
-                # data = data.to_html()
                 json_records = df.reset_index().to_json(orient='records')
                 arr = []
                 arr = json.loads(json_records)
                 if (arr == [] ):
-                    # data = data.to_html()
                     json_records = df.reset_index().to_json(orient='records')
                     arr = []
                     arr = json.loads(json_records)
@@ -908,11 +904,8 @@ def NextMaintain(request):
                 form = MyFormServices()
 
                 enddate = enddate.strftime("%d/%m/%Y")
-               # parts = enddate.split(' ')
-               # message_count = int(parts[0])
-             #   print(message_count)
 
-    #ARIMA SECTION
+#Price Prediction with ARIMA
 
                 df2 = pd.read_csv('Data/MaintainData.csv')
                 df2 = df2[(df2["Customer"] == customerid)]
@@ -927,11 +920,6 @@ def NextMaintain(request):
                 ## Cleaning up the data
                 df.columns = ["Date", "Price"]
                 rrr = df.head()
-                ## Drop last 2 rows
-                #    df.drop(106, axis=0, inplace=True)
-                #  df.tail()
-                #  df.drop(105, axis=0, inplace=True)
-                #   df.tail()
 
                 # Convert Date into Datetime
                 df['Date'] = pd.to_datetime(df['Date'])
@@ -941,7 +929,7 @@ def NextMaintain(request):
                 df.describe()
                 # df.plot()
                 # plt.show()
-                ### Testing For Stationarity
+                ##Testing For Stationarity
 
                 from statsmodels.tsa.stattools import adfuller
                 test_result = adfuller(df['Price'])
@@ -959,31 +947,23 @@ def NextMaintain(request):
                             "weak evidence against null hypothesis, time series has a unit root, indicating it is non-stationary ")
 
                 adfuller_test(df['Price'])
-                ##Differencing
+                #Differencing
                 df['Price First Difference'] = df['Price'] - df['Price'].shift(1)
                 df['Price'].shift(1)
                 df['Seasonal First Difference'] = df['Price'] - df['Price'].shift(12)
-                # df.head(14)
 
-                ## Again test dickey fuller test
+                #Again test dickey fuller test
                 adfuller_test(df['Seasonal First Difference'].dropna())
                 import statsmodels.api as sm
-                # For non-seasonal data
-                # p=1, d=1, q=0 or 1
-                #   from statsmodels.tsa.arima_model import ARIMA
                 from statsmodels.tsa.arima.model import ARIMA
                 model = ARIMA(df['Price'], order=(1, 1, 1))
                 model_fit = model.fit()
-                # model_fit.summary()
-                #       df['forecast'] = model_fit.predict(start=8, end=8, dynamic=True)
-                # df[['Price','forecast']].plot(figsize=(12,8))
                 model = sm.tsa.statespace.SARIMAX(df['Price'], order=(1, 1, 1), seasonal_order=(1, 1, 1, 12))
                 results = model.fit()
 
                 from pandas.tseries.offsets import DateOffset
                 future_dates = [df.index[-1] + DateOffset(months=x) for x in range(0, 150)]
                 future_datest_df = pd.DataFrame(index=future_dates[1:], columns=df.columns)
-                # future_datest_df.tail()
                 future_df = pd.concat([df, future_datest_df])