Create app.py

app.py

@@ -0,0 +1,99 @@
+# import required libraries
+
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+import pickle
+
+from datetime import datetime
+from datetime import timedelta
+from sklearn.model_selection import RandomizedSearchCV, GridSearchCV, train_test_split
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.metrics import r2_score
+from sklearn.preprocessing import LabelEncoder
+from sklearn.preprocessing import StandardScaler
+import streamlit as st 
+import warnings
+warnings.filterwarnings('ignore')
+
+st.title("Predict Unrolled Values")
+st.sidebar.header('Enter the Details here')
+st.write("""Regression Model""")
+
+# load the saved model using pickle
+with open('aajTak_model.pkl', 'rb') as f1:
+    model1 = pickle.load(f1)
+    
+# Load the saved weekDay label encoder object using pickle
+with open('weekDay_le.pkl','rb') as file1:
+    weekDay_le = pickle.load(file1)
+
+# Load the saved timeBand label encoder object using pickle
+with open('timeBand_le.pkl','rb') as file2:
+    timeBand_le = pickle.load(file2)
+
+# previous_number_of_repairs = 
+# st.sidebar.number_input('Enter the Previous Number of Repairs Undergone  0 to 5 )',min_value=0,max_value=5,step=1)  
+    
+# DATA from user
+def user_report():
+  
+  Share = round(float(st.sidebar.slider('Share', 0.000000, 100.000000, 0.611246)), 6)
+  AMA = round(float(st.sidebar.slider('AMA', 0.000000, 45.000000, 4.196084)), 6)
+  rate = round(float(st.sidebar.slider('rate', 0.000000, 1.500000, 0.018516)), 6)  
+  daily_reach = round(float(st.sidebar.slider('daily reach', 0.000000, 300.000000, 36.23)), 6) 
+  cume_reach = round(float(st.sidebar.slider('cume reach', 0.000000, 300.000000, 36.231006)), 6)
+    
+  # Output: {'Friday': 0, 'Monday': 1, 'Saturday': 2, 'Sunday': 3, 'Thursday': 4, 'Tuesday': 5, 'Wednesday': 6}
+    
+  Week_Day_Encoded = st.sidebar.selectbox("Week Day",
+        ("Monday", "Tuesday","Wednesday","Thursday","Friday", "Saturday", "Sunday" ))
+  if Week_Day_Encoded=='Monday':
+      Week_Day_Encoded=1
+  elif Week_Day_Encoded=="Tuesday":
+      Week_Day_Encoded=5
+  elif Week_Day_Encoded=="Wednesday":
+      Week_Day_Encoded=6
+  elif Week_Day_Encoded=="Thursday":
+      Week_Day_Encoded =4
+  elif Week_Day_Encoded=="Friday":
+      Week_Day_Encoded =0
+  elif Week_Day_Encoded=="Saturday":
+      Week_Day_Encoded =2
+  else:
+      Week_Day_Encoded=3
+  
+  user_report_data = {
+       'Share': Share, 
+       'AMA': AMA, 
+       'rate': rate, 
+       'daily reach': daily_reach, 
+       'cume reach': cume_reach, 
+       'Week_Day_Encoded': Week_Day_Encoded, 
+       'Time_Band_Encoded': Time_Band_Encoded}
+  report_data = pd.DataFrame(user_report_data, index=[0])
+  
+  return report_data    
+    
+#Customer Data
+user_data = user_report()
+st.subheader("Entered Details")
+st.write(user_data)
+
+
+# define the prediction function
+def predict_unrolled_value(user_data):
+    
+    # make the prediction using the loaded model and input data
+    predicted_unrolled_value = model1.predict(user_data)
+    
+    # return the predicted max number of repairs as output
+    return np.round(predicted_unrolled_value[0])
+
+    
+# Function calling
+y_pred = int(predict_unrolled_value(user_data))
+st.write("Click here to see the Predictions")
+if st.button("Predict"):
+    st.subheader(f"Predicted Unrolled Value: {y_pred} ")