#!/usr/bin/env python

# coding: utf-8

# In[1]:

import numpy as np

import pandas as pd

# In[2]:

d=pd.read_csv(r'D:\HousePredictions\train.csv',encoding='unicode_escape')

d.head()

# In[3]:

d.describe()

# In[4]:

d.info()

# In[5]:

d_missing=d.isna().sum()

# missing[d_missing>0].sort_values(ascending=False)

# In[6]:

d_missing[d_missing>0].sort_values(ascending=False)

# In[7]:

#keeping only columns which dont have na

# In[8]:

d=d.dropna(axis=1, how='any')

d.shape

# In[9]:

#removing Id field which doesnt have impact on house price.

#del d['Id']

d.head()

# In[10]:

#corelation matrix

# In[86]:

import seaborn as sns

import matplotlib.pyplot as plt

matrix = d.corr()

f, ax = plt.subplots(figsize=(16, 12))

sns.heatmap(matrix, vmax=0.7, square=True)

# In[87]:

#selcting only features which are highly correlated

tcf=matrix['SalePrice'].sort_values(ascending=False)

# In[88]:

# Filter out the target variables (SalePrice) and variables with a low correlation score (v such that -0.6 <= v <= 0.6)

tcf = tcf[abs(tcf) >= 0.6]

# In[89]:

tcf = tcf[tcf.index != 'SalePrice']

tcf

# In[90]:

tcf.shape

# In[91]:

cols = tcf.index.values.tolist() + ['SalePrice']

sns.pairplot(d[cols], size=2.5)

plt.show()

# In[94]:

# Build the correlation matrix

matrix = d[cols].corr()

f, ax = plt.subplots(figsize=(8, 6))

sns.heatmap(matrix, vmax=1.0, square=True)

# In[146]:

from sklearn.ensemble import RandomForestClassifier

from sklearn.model_selection import train_test_split

X=d1.loc[:,d1.columns!='SalePrice']

y = d1['SalePrice']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

model = RandomForestClassifier(n_estimators=100, random_state=42)

model.fit(X_train, y_train)

# In[147]:

y_pred = model.predict(X_test)

# Build a plot

plt.scatter(y_pred, y_test)

plt.xlabel('Prediction')

plt.ylabel('Real value')

# Now add the perfect prediction line

diagonal = np.linspace(0, np.max(y_test), 100)

plt.plot(diagonal, diagonal, '-r')

plt.show()

# In[148]:

from sklearn.metrics import mean_squared_log_error, mean_absolute_error

print('MAE:\t$%.2f' % mean_absolute_error(y_test, y_pred))

print('MSLE:\t%.5f' % mean_squared_log_error(y_test, y_pred))

# In[149]:

#Score/Accuracy

print("Accuracy --> ", model.score(X_test, y_test)*100)

# In[153]:

#Train the model

from sklearn import linear_model

model = linear_model.LinearRegression()

# In[154]:

#Fit the model

model.fit(X_train, y_train)

# In[157]:

#Score/Accuracy

print("Accuracy --> ", model.score(X_test, y_test)*100)

# In[159]:

# Build a plot

plt.scatter(y_pred, y_test)

plt.xlabel('Prediction')

plt.ylabel('Real value')

# Now add the perfect prediction line

diagonal = np.linspace(0, np.max(y_test), 100)

plt.plot(diagonal, diagonal, '-r')

plt.show()

# In[150]:

#Train the model

from sklearn.ensemble import GradientBoostingRegressor

GBR = GradientBoostingRegressor(n_estimators=100, max_depth=4)

# In[151]:

#Fit

GBR.fit(X_train, y_train)

# In[152]:

print("Accuracy --> ", GBR.score(X_test, y_test)*100)

# In[158]:

# Build a plot

plt.scatter(y_pred, y_test)

plt.xlabel('Prediction')

plt.ylabel('Real value')

# Now add the perfect prediction line

diagonal = np.linspace(0, np.max(y_test), 100)

plt.plot(diagonal, diagonal, '-r')

plt.show()

# In[ ]: