Update app.py

app.py

@@ -1,3 +1,12 @@
+# -*- coding: utf-8 -*-
+"""car_price.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1rtMdlilQhGBozNcdxDeSkuEthtwAz7-L
+"""
+
 import pandas as pd
 import numpy as np
 import warnings
@@ -25,34 +34,40 @@ from sklearn.model_selection import train_test_split
 def train_test(df):
         X=df.drop('Price',axis=1)
         y=df['Price']
-        X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.15,random_state=43)
+        X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=43)
         return X_train,X_test,y_train,y_test
 
-df = datacleanning(df)
-X_train,X_test,y_train,y_test = train_test(df)
+def dummie(X_train,X_test):
+    X_train = pd.get_dummies(X_train,drop_first=True)
+    X_test = pd.get_dummies(X_test,drop_first=True)
+    return X_train,X_test
+
+def final_df(df):
+  df =datacleanning(df)
+  X_train,X_test,y_train,y_test = train_test(df)
+  X_train,X_test = dummie(X_train,X_test)
+  return df,X_train,X_test,y_train,y_test
 
-from sklearn.pipeline import Pipeline
-from sklearn.compose import ColumnTransformer
-from sklearn.preprocessing import OneHotEncoder
-from sklearn.preprocessing import StandardScaler
 from sklearn.ensemble import HistGradientBoostingRegressor
+def model_fit(X_train,y_train):
+  hgb = HistGradientBoostingRegressor()
+  hgb.fit(X_train,y_train)
+  return hgb
+
+model = model_fit(X_train,y_train)
 
-hgbr = HistGradientBoostingRegressor()
-numeric_features = ['Model Year', 'Mileage', 'Engine Capacity']
-categorical_features = ['Company Name', 'Model Name', 'Location', 'Engine Type', 'Color', 'Assembly', 'Body Type', 'Transmission Type', 'Registration Status']
-transformer = ColumnTransformer(transformers=[('num',StandardScaler(),numeric_features),('cat',OneHotEncoder(drop='if_binary', dense=True),categorical_features)])
-pipe = Pipeline(steps=[('preprocessor', transformer), ('regressor', hgbr)])
+!pip install streamlit
 
-pipe.fit(X_train,y_train)
+df = pd.get_dummies(df,drop_first=True)
 
 import streamlit as st
 
-def price(companyName,modelName,modelYear,location,mileage,engineType,engineCapacity,color,assembly,bodyType,transmissionType,registrationStatus):
+def price(companyName,modelName,modelYear,locaiton,mileage,engineType,engineCapacity,color,assembly,bodyType,transmissionType,registrationStatus):
   input_data = pd.DataFrame({
       'Company Name':[companyName],
       'Model Name':[modelName],
       'Model Year':[modelYear],
-      'Location':[location],
+      'Location':[locaiton],
       'Mileage':[mileage],
       'Engine Type':[engineType],
       'Engine Capacity':[engineCapacity],
@@ -60,27 +75,25 @@ def price(companyName,modelName,modelYear,location,mileage,engineType,engineCapa
       'Assembly':[assembly],
       'Body Type':[bodyType],
       'Transmission Type':[transmissionType],
-      'Registration Status':[registrationStatus == 'Yes']
+      'Registration Status':[registrationStatus]
   })
-  input_data = datacleanning(input_data)
-  prediction=pipe.predict(input_data)[0]
+  prediction=model.predict(input_data)[0]
   return prediction
-
 st.title('Car Price Prediction:car @yusufenes')
 st.write('Please Chose Car Specifications')
 companyName = st.selectbox('Company Name',df['Company Name'].unique())
 modelName = st.selectbox('Model Name',df[df['Company Name']==companyName]['Model Name'].unique())
 modelYear = st.selectbox('Model Year',df[(df['Company Name']==companyName)&(df['Model Name'] == modelName)]['Model Year'].unique())
-location = st.selectbox('Location',df['Location'].unique())
+locaiton = st.selectbox('Location',df['Location'].unique())
 mileage = st.number_input('Mileage',df['Mileage'].min(),df['Mileage'].max())
 engineType = st.selectbox('Engine Type',df['Engine Type'].unique())
 engineCapacity = st.number_input('Engine Capacity',df['Engine Capacity'].min(),df['Engine Capacity'].max())
 color = st.selectbox('Color',df['Color'].unique())
 assembly = st.selectbox('Assembly',df['Assembly'].unique())
 bodyType = st.selectbox('Body Type',df['Body Type'].unique())
-transmissionType= st.selectbox('Transmission Type',df['Transmission Type'].unique())
+transmissionType = st.selectbox('Transmission Type',df['Transmission Type'].unique())
 registrationStatus = st.radio('Registration Status',['Yes','No'])
 if st.button('Predict'):
-  pred=price(companyName,modelName,modelYear,location,mileage,engineType,engineCapacity,color,assembly,bodyType,transmissionType,registrationStatus)
+  pred=price(companyName,modelName,modelYear,locaiton,mileage,engineType,engineCapacity,color,assembly,bodyType,transmissionType,registrationStatus)
   st.success(f'The predicted price is {pred} $')
-  st.balloons()
+  st.balloons()