ThirdEyeData/Component_Repair_Time_Prediction

1import tensorflow as tf2from tensorflow import keras3import numpy as np4import matplotlib.pyplot as plt5import pandas as pd6from sklearn.model_selection import train_test_split7from sklearn import preprocessing8import seaborn as sns9from sklearn.preprocessing import LabelEncoder10import pickle11import streamlit as st 12 13st.title('Repair Time Prediction')14#DLoading the ataset15#df = pd.read_csv('repair_time_sample_50k_modified2.csv')16 17#new_data = df18#df.drop(['SRU serial number','Date of Manufacture', 'Snag Description'], axis = 1, inplace=True)19 20 21# DATA from user22def user_report():23  Aircraft_Type = st.sidebar.selectbox('Aircraft Type',("AH-64","UH-60","UH-63","UH-62","UH-61","AH-65"))24  if Aircraft_Type=="AH-64":25      Aircraft_Type=026  elif Aircraft_Type=="UH-60":27      Aircraft_Type=228  elif Aircraft_Type=="UH-63":29      Aircraft_Type=530  elif Aircraft_Type=="UH-62":31      Aircraft_Type=432  elif Aircraft_Type=="UH-61":33      Aircraft_Type=334  else:35      Aircraft_Type=136  manufacturer = st.sidebar.selectbox("Manufacturer",37        ("JKL Company", "GHI Company","AGS Company","ABC Company","XYZ Company" ))38  if manufacturer=='JKL Company':39      manufacturer=340  elif manufacturer=="GHI Company":41      manufacturer=242  elif manufacturer=="AGS Company":43      manufacturer=144  elif manufacturer=="ABC Company":45      manufacturer =046  else:47      manufacturer=448  component_age = st.sidebar.slider('Component Age (in hours)', 500,2000, 600 )49  Issue_category= st.sidebar.selectbox("Issue Category",50        ("Display", "Unservicable","Bootup Problem","Engine Failure","Electrical Fault" ))51  if Issue_category=='Display':52      Issue_category=153  elif Issue_category=="Unservicable":54      Issue_category=455  elif Issue_category=="Bootup Problem":56      Issue_category=057  elif Issue_category=="Engine Failure":58      Issue_category=359  else:60      Issue_category=261  Snag_Severity	= st.sidebar.selectbox("Snag Severity",62        ("Low", "Medium","High" ))63  if Snag_Severity	=='Low':64      Snag_Severity=165  elif Snag_Severity=="Medium":66      Snag_Severity	=267  else:68       Snag_Severity=069  Customer= st.sidebar.selectbox("Customer",70        ("IAF", "ARMY","NAVY" ))71  if Customer	=='IAF':72      Customer=173  elif Customer=="ARMY":74      Customer	=075  else:76       Customer=277  Technician_Skill_level= st.sidebar.selectbox("Technician Skill level",78        ("Expert", "Intermediate","Novice" ))79  if Technician_Skill_level	=='Expert':80      Technician_Skill_level=081  elif Technician_Skill_level=="Intermediate":82      Technician_Skill_level	=183  else:84       Technician_Skill_level=285  prior_maintainence = st.sidebar.selectbox('Prior Maintainence',("Regular","Irregular"))86  if prior_maintainence =='Regular':87      prior_maintainence=188  else:89      prior_maintainence=090  Logistics_Time = st.sidebar.slider('Logistics Time (hr)', 2,21, 5 )91  total_operating_hours = st.sidebar.slider('Total Operating Hours)', 50,2000, 500 )92  operating_temperature = st.sidebar.slider('Operating Temperature', 10,25, 15 )93  previous_number_of_repairs = st.sidebar.number_input('Enter the Previous Number of Repairs Undergone  0 to 3 )',min_value=0,max_value=3,step=1)94  Power_Input_Voltage= st.sidebar.slider('Power Input Voltage (V)',100,133,115)  95 96 97  user_report_data = {98      'Aircraft Type':Aircraft_Type,99      'Manufacturer':manufacturer,100      'Component_Age':component_age,101      'Issue_category':Issue_category,102      'Snag Severity': Snag_Severity,103      'Customer':Customer,104      'Technician Skill level':Technician_Skill_level,105      'Prior Maintenance': prior_maintainence,106      'Logistics Time (hr)':Logistics_Time,107      'total_operating_hours':total_operating_hours,108      'operating_temperature':operating_temperature,109      'previous_number_of_repairs':previous_number_of_repairs,110      'Power_Input_Voltage':Power_Input_Voltage111           112  }113  report_data = pd.DataFrame(user_report_data, index=[0])114  return report_data    115    116#Customer Data117user_data = user_report()118st.header("Component Details")119st.write(user_data)120 121def preprocess_dataset(X):  122  x = X.values #returns a numpy array123  min_max_scaler = preprocessing.MinMaxScaler()124  x_scaled = min_max_scaler.fit_transform(x)125  X_df = pd.DataFrame(x_scaled)126  return X_df127 128def label_encoding(data):129  le = LabelEncoder()130  cat = data.select_dtypes(include='O').keys()131  categ = list(cat)132  data[categ] = data[categ].apply(le.fit_transform)133  # X = data.loc[:,data.columns!= "Time required for repair (in hours)"]134  # y = data['Time required for repair (in hours)']135  # return X,y136  return data137 138def prediction(df):139  #X = df.loc[:,df.columns!= "Time required for repair (in hours)"]140  #y = df['Time required for repair (in hours)']141  #X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)142  #print(X_train.shape)143  #print(X_test.shape)144  #X_test_encoded = label_encoding(df)145  #X_test_df = preprocess_dataset(df)146  x_model = pickle.load(open('repair_time_model.pkl','rb'))147  pred = x_model.predict(df)148  #X_test['Actual_time_to_repair'] = y_test149  #X_test['Predicted_time_to_repair'] = pred150  #X_test.to_csv(r'/content/drive/MyDrive/Colab Notebooks/HAL/repair_time_prediction_results.csv')151  #print(X_test.head())152  return pred153    154y_pred = prediction(user_data)155 156if st.button("Predict"):157    st.subheader(f"Time required to Repair the Component is {y_pred[0]} hours")