Hugging Face's logo

Spaces:
ayethuzar
/
HousePricePredictionApp
Sleeping

App Files Community
ayethuzar commited on
Commit
861644b
·
unverified ·
1 Parent(s): 8699775

app update

Browse files
Files changed (1) hide show
  1. app.py +217 -0
app.py ADDED
@@ -0,0 +1,217 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+
3
+ import streamlit as st
4
+ import pandas as pd
5
+ import numpy as np
6
+ import tensorflow as tf
7
+
8
+ from sklearn.preprocessing import MinMaxScaler
9
+
10
+ import tensorflow_probability as tfp
11
+ tfd = tfp.distributions
12
+
13
+ from pickle import dump
14
+ from pickle import load
15
+
16
+
17
+
18
+ scaler = MinMaxScaler()
19
+
20
+
21
+ # load trained model
22
+ lgbm_base = pickle.load(open('lgbm_base.pkl', 'rb'))
23
+ lgbm_opt = pickle.load(open('lgbm_optimized.pkl', 'rb'))
24
+
25
+
26
+ tf.random.set_seed(42)
27
+
28
+ np.random.seed(42)
29
+
30
+
31
+ st.markdown("<body style ='color:#E2E0D9;'></body>", unsafe_allow_html=True)
32
+
33
+
34
+
35
+ st.markdown("<h4 style='text-align: center; color: #1B9E91;'>House Price Prediction in Ames,Iowa</h4>", unsafe_allow_html=True)
36
+
37
+ st.markdown("<h5 style='text-align: center; color: #1B9E91;'>A multi-step process is used to estimate the range of house prices based on your selection. </h5>", unsafe_allow_html=True)
38
+
39
+
40
+ name_list = [
41
+ 'OverallQual',
42
+ 'YearBuilt',
43
+ 'TotalBsmtSF',
44
+ 'GrLivArea',
45
+ 'MasVnrArea',
46
+ 'BsmtFinType1',
47
+ 'Neighborhood',
48
+ 'GarageType',
49
+ 'SaleCondition',
50
+ 'BsmtExposure']
51
+
52
+ name_list_train = [
53
+ 'OverallQual',
54
+ 'YearBuilt',
55
+ 'TotalBsmtSF',
56
+ 'GrLivArea',
57
+ 'MasVnrArea',
58
+ 'BsmtFinType1',
59
+ 'Neighborhood',
60
+ 'GarageType',
61
+ 'SaleCondition',
62
+ 'BsmtExposure']
63
+
64
+ data = pd.read_csv('train.csv')
65
+
66
+
67
+ data = data[name_list_train].values
68
+
69
+ scaler.fit(data)
70
+
71
+ description_list = [
72
+ 'What is the Overall material and finish quality?',
73
+ 'In which year was the Original construction date?',
74
+ 'What is the Total square feet of basement area?',
75
+ 'What is the Above grade (ground) living area in square feet?',
76
+ 'What is the Masonry veneer area in square feet??',
77
+ 'What is the Quality of basement finished area?',
78
+ 'Where is the physical locations within Ames city limits?',
79
+ 'Where is the location of the Garage?',
80
+ 'What is the condition of the sale?',
81
+ 'Does the house have walkout or garden-level basement walls?'
82
+ ]
83
+
84
+ min_list = [1.0,
85
+ 1950.0,
86
+ 0.0,
87
+ 0.0,
88
+ 334.0,
89
+ 0.0,
90
+ 0.0,
91
+ 0.0,
92
+ 0.0,
93
+ 0.0
94
+ ]
95
+
96
+ max_list = [
97
+ 10.0,
98
+ 2010.0,
99
+ 2336.0,
100
+ 6110.0,
101
+ 4692.0,
102
+ 10.0,
103
+ 10.0,
104
+ 3.0,
105
+ 10.0,
106
+ 1.0,
107
+ ]
108
+
109
+ count = 0
110
+
111
+ with st.sidebar:
112
+
113
+ for i in range(len(name_list)):
114
+
115
+
116
+
117
+ variable_name = name_list[i]
118
+ globals()[variable_name] = st.slider(description_list[i] ,min_value=int(min_list[i]), max_value =int(max_list[i]),step=1)
119
+
120
+ st.write("[Kaggle Link to Data Set](https://www.kaggle.com/competitions/house-prices-advanced-regression-techniques)")
121
+
122
+
123
+
124
+
125
+ data_df = {
126
+
127
+ 'OverallQual': [OverallQual],
128
+ 'YearBuilt': [YearBuilt],
129
+ 'TotalBsmtSF': [TotalBsmtSF],
130
+ 'GrLivArea':[GrLivArea],
131
+ 'MasVnrArea': [MasVnrArea]
132
+ 'BsmtFinType1': [BsmtFinType1]
133
+ 'Neighborhood': [Neighborhood]
134
+ 'GarageType': [GarageType]
135
+ 'SaleCondition': [SaleCondition]
136
+ 'BsmtExposure': [BsmtExposure]
137
+ }
138
+
139
+ #negloglik = lambda y, p_y: -p_y.log_prob(y) # note this
140
+
141
+ data_df = pd.DataFrame.from_dict(data_df)
142
+
143
+ data_df_normal = scaler.transform(data_df)
144
+
145
+ y_pred_base = lgbm_base.predict(data_df)
146
+ y_pred_optimized = lgbm_opt.predict(data_df)
147
+
148
+
149
+
150
+ col1, col2, col3 , col4, col5 = st.columns(5)
151
+
152
+ with col1:
153
+ pass
154
+ with col2:
155
+ pass
156
+ with col4:
157
+ pass
158
+ with col5:
159
+ pass
160
+ with col3 :
161
+ center_button = st.button('Calculate range of house price')
162
+
163
+
164
+
165
+ if center_button:
166
+
167
+ import time
168
+
169
+ #my_bar = st.progress(0)
170
+
171
+ with st.spinner('Calculating....'):
172
+ time.sleep(2)
173
+
174
+
175
+
176
+ st.markdown("<h5 style='text-align: center; color: #1B9E91;'>The price range of your house is between:</h5>", unsafe_allow_html=True)
177
+
178
+
179
+ col1, col2 = st.columns([3, 3])
180
+
181
+ lower_number = "{:,.2f}".format(int(yhat.mean().numpy()-1.95*yhat.stddev().numpy()))
182
+ higher_number = "{:,.2f}".format(int(yhat.mean().numpy()+1.95*yhat.stddev().numpy()))
183
+
184
+ col1, col2, col3 = st.columns(3)
185
+
186
+
187
+
188
+ with col1:
189
+ st.write("")
190
+
191
+ with col2:
192
+ st.subheader("USD "+ str(lower_number))
193
+ st.subheader(" AND ")
194
+
195
+ st.subheader(" USD "+str(higher_number))
196
+
197
+
198
+ with col3:
199
+ st.write("")
200
+
201
+
202
+
203
+
204
+
205
+ import base64
206
+
207
+ file_ = open("kramer_gif.gif", "rb")
208
+ contents = file_.read()
209
+ data_url = base64.b64encode(contents).decode("utf-8")
210
+ file_.close()
211
+
212
+ st.markdown(
213
+ f'<center><img src="data:image/gif;base64,{data_url}" alt="cat gif"></center>',
214
+ unsafe_allow_html=True,
215
+ )
216
+
217
+