Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,239 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import pickle
|
3 |
+
import pandas as pd
|
4 |
+
from sentence_transformers import SentenceTransformer, util
|
5 |
+
import numpy as np
|
6 |
+
import urllib.parse
|
7 |
+
import requests
|
8 |
+
|
9 |
+
#initial state
|
10 |
+
if 'state_p1' not in st.session_state:
|
11 |
+
st.session_state.state_p1 = 0
|
12 |
+
if 'state_p2' not in st.session_state:
|
13 |
+
st.session_state.state_p2 = 0
|
14 |
+
if 'age' not in st.session_state:
|
15 |
+
st.session_state.age = 0
|
16 |
+
if 'weight' not in st.session_state:
|
17 |
+
st.session_state.weight = 0
|
18 |
+
if 'height' not in st.session_state:
|
19 |
+
st.session_state.height = 0
|
20 |
+
if 'gender' not in st.session_state:
|
21 |
+
st.session_state.gender = 0
|
22 |
+
if 'food_allergy' not in st.session_state:
|
23 |
+
st.session_state.food_allergy = 0
|
24 |
+
if 'drug_allergy' not in st.session_state:
|
25 |
+
st.session_state.drug_allergy = 0
|
26 |
+
if 'congentital_disease' not in st.session_state:
|
27 |
+
st.session_state.congentital_disease = 0
|
28 |
+
if 'queries' not in st.session_state:
|
29 |
+
st.session_state.queries = None
|
30 |
+
if 'sbert_searched_df' not in st.session_state:
|
31 |
+
st.session_state.sbert_searched_df = None
|
32 |
+
if 'queries_p2' not in st.session_state:
|
33 |
+
st.session_state.queries_p2 = None
|
34 |
+
if 'sbert_searched_df_p2' not in st.session_state:
|
35 |
+
st.session_state.sbert_searched_df_p2 = None
|
36 |
+
for i in range(10):
|
37 |
+
if 'score_'+str(i+1) not in st.session_state:
|
38 |
+
st.session_state['score_'+str(i+1)] = 'NA'
|
39 |
+
if 'current_page' not in st.session_state:
|
40 |
+
st.session_state.current_page = 1
|
41 |
+
|
42 |
+
def set_state_p1(state):
|
43 |
+
st.session_state.state_p1 = state
|
44 |
+
|
45 |
+
def set_state_p2(state):
|
46 |
+
st.session_state.state_p2 = state
|
47 |
+
|
48 |
+
def split_text(text):
|
49 |
+
return text.split(',')
|
50 |
+
|
51 |
+
#import data
|
52 |
+
sbert_model = SentenceTransformer('paraphrase-multilingual-mpnet-base-v2')
|
53 |
+
|
54 |
+
data = pd.read_csv('articles_data.csv')
|
55 |
+
|
56 |
+
with open('corpus_embeddings.pickle', 'rb') as file:
|
57 |
+
corpus_embeddings = pickle.load(file)
|
58 |
+
|
59 |
+
#local function
|
60 |
+
def personal_check(age,weight,height,gender):
|
61 |
+
|
62 |
+
#age check
|
63 |
+
if age >= 60:
|
64 |
+
age = 'ผู้สูงอายุ'
|
65 |
+
else:
|
66 |
+
age = 'ทำงาน'
|
67 |
+
|
68 |
+
#gender check
|
69 |
+
if gender == 'หญิง':
|
70 |
+
gender = 'ผู้หญิง'
|
71 |
+
else:
|
72 |
+
gender = 'ผู้ชาย'
|
73 |
+
|
74 |
+
#bmi check
|
75 |
+
height_meters = height / 100
|
76 |
+
|
77 |
+
bmi = weight / (height_meters ** 2)
|
78 |
+
|
79 |
+
if bmi >= 30:
|
80 |
+
bmi = 'อ้วนมาก'
|
81 |
+
elif bmi >= 23 and bmi <30:
|
82 |
+
bmi = 'อ้วน'
|
83 |
+
elif bmi >= 18.5 and bmi <23:
|
84 |
+
bmi = ''
|
85 |
+
else:
|
86 |
+
bmi = 'ผอม'
|
87 |
+
|
88 |
+
return age,gender,bmi
|
89 |
+
|
90 |
+
def sbert_search(queries,data,embeddiing,sbert_model=sbert_model):
|
91 |
+
|
92 |
+
index_lst = []
|
93 |
+
score_lst = []
|
94 |
+
|
95 |
+
query_embedding = sbert_model.encode(queries, convert_to_tensor=True)
|
96 |
+
hits = util.semantic_search(query_embedding, embeddiing, top_k=10)
|
97 |
+
hits = hits[0]
|
98 |
+
for hit in hits:
|
99 |
+
index_lst.append(hit['corpus_id'])
|
100 |
+
score_lst.append(hit['score'])
|
101 |
+
|
102 |
+
sbert_searched = data.iloc[index_lst]
|
103 |
+
sbert_searched['score'] = score_lst
|
104 |
+
|
105 |
+
return sbert_searched
|
106 |
+
|
107 |
+
def page1_recommendation():
|
108 |
+
#header
|
109 |
+
st.markdown("<h1 style='text-align: center; color: black;'>---ระบบแนะนำบทความสุขภาพ---</h1>", unsafe_allow_html=True)
|
110 |
+
|
111 |
+
with st.form('user_info'):
|
112 |
+
|
113 |
+
#personal information input
|
114 |
+
age = st.slider("อายุ", 10, 100, 25)
|
115 |
+
|
116 |
+
col1, col2 = st.columns(2)
|
117 |
+
with col1:
|
118 |
+
weight = st.number_input("น้ำหนัก (Kg.): ",30.0,120.0,step=1.0,value=50.0)
|
119 |
+
with col2:
|
120 |
+
height = st.number_input("ส่วนสูง (cm.): ",100.0,250.0,step=1.0,value=150.0)
|
121 |
+
|
122 |
+
col3, col4, col5 = st.columns(3)
|
123 |
+
with col3:
|
124 |
+
gender = st.selectbox('เพศ',('ชาย', 'หญิง'))
|
125 |
+
with col4:
|
126 |
+
food_allergy = st.selectbox('แพ้อาหาร?',('ไม่แพ้', 'แพ้อาหาร'))
|
127 |
+
with col5:
|
128 |
+
drug_allergy = st.selectbox('แพ้ยา?',('ไม่แพ้', 'แพ้ยา'))
|
129 |
+
congentital_disease = st.text_input('โรคประจำตัวของคุณ (ถ้าหากไม่มี ไม่ต้องกรอก หรือใส่ "ไม่มี")')
|
130 |
+
|
131 |
+
st.form_submit_button(on_click=set_state_p1,args=(1,))
|
132 |
+
|
133 |
+
if st.session_state.state_p1 == 1:
|
134 |
+
|
135 |
+
#asign state
|
136 |
+
st.session_state.age = age
|
137 |
+
st.session_state.weight = weight
|
138 |
+
st.session_state.height = height
|
139 |
+
st.session_state.gender = gender
|
140 |
+
st.session_state.food_allergy = food_allergy
|
141 |
+
st.session_state.drug_allergy = drug_allergy
|
142 |
+
st.session_state.congentital_disease = congentital_disease
|
143 |
+
|
144 |
+
#algorithm
|
145 |
+
age,gender,bmi = personal_check(age,weight,height,gender)
|
146 |
+
|
147 |
+
if food_allergy == 'ไม่แพ้':
|
148 |
+
food_allergy = ''
|
149 |
+
if drug_allergy == 'ไม่แพ้':
|
150 |
+
drug_allergy = ''
|
151 |
+
if congentital_disease == 'ไม่มี':
|
152 |
+
congentital_disease = ''
|
153 |
+
|
154 |
+
if congentital_disease == '':
|
155 |
+
queries = gender+age+bmi+food_allergy+drug_allergy
|
156 |
+
else:
|
157 |
+
queries = congentital_disease
|
158 |
+
|
159 |
+
#Bertopic search
|
160 |
+
sbert_searched = sbert_search(queries,data,corpus_embeddings)
|
161 |
+
|
162 |
+
st.session_state.sbert_searched_df = sbert_searched
|
163 |
+
st.session_state.queries = queries
|
164 |
+
st.session_state.state_p1 = 2
|
165 |
+
|
166 |
+
if st.session_state.state_p1 == 2:
|
167 |
+
|
168 |
+
with st.form('recommendations'):
|
169 |
+
st.markdown("<h1 style='text-align: center; color: black;'>📰บทความสำหรับคุณ😆</h1>", unsafe_allow_html=True)
|
170 |
+
st.write("---------------------------------------------------------------------------------------")
|
171 |
+
|
172 |
+
for i in range(len(st.session_state.sbert_searched_df)):
|
173 |
+
st.header(str(i+1)+'. '+st.session_state.sbert_searched_df.iloc[i]['title'])
|
174 |
+
st.markdown('**Keywords :** '+ st.session_state.sbert_searched_df.iloc[i]['vote_keywords'])
|
175 |
+
st.markdown(f"[Page source (Click here.)]({st.session_state.sbert_searched_df.iloc[i].url})")
|
176 |
+
|
177 |
+
try:
|
178 |
+
banner_url = urllib.parse.quote(st.session_state.sbert_searched_df.iloc[i]['banner'], safe=':/')
|
179 |
+
response = requests.get(banner_url,timeout=5)
|
180 |
+
st.image(response.content)
|
181 |
+
except:
|
182 |
+
st.image('https://icon-library.com/images/no-photo-icon/no-photo-icon-1.jpg')
|
183 |
+
finally:
|
184 |
+
st.write("---------------------------------------------------------------------------------------")
|
185 |
+
|
186 |
+
st.form_submit_button('Submit',on_click=set_state_p1,args=(0,))
|
187 |
+
|
188 |
+
def page2_search_engine():
|
189 |
+
st.title("Search engine")
|
190 |
+
|
191 |
+
with st.form('queries'):
|
192 |
+
queries = st.text_input('คำหรือหัวข้อที่ต้องการค้นหา')
|
193 |
+
st.form_submit_button(on_click=set_state_p2,args=(1,))
|
194 |
+
|
195 |
+
if st.session_state.state_p2 == 1:
|
196 |
+
sbert_searched = sbert_search(queries,data,corpus_embeddings)
|
197 |
+
|
198 |
+
st.session_state.sbert_searched_df_p2 = sbert_searched
|
199 |
+
st.session_state.queries_p2 = queries
|
200 |
+
st.session_state.state_p2 = 2
|
201 |
+
|
202 |
+
if st.session_state.state_p2 == 2:
|
203 |
+
with st.form('recommendations'):
|
204 |
+
st.markdown("<h1 style='text-align: center; color: black;'>📰บทความสำหรับคุณ😆</h1>", unsafe_allow_html=True)
|
205 |
+
st.write("---------------------------------------------------------------------------------------")
|
206 |
+
|
207 |
+
for i in range(len(st.session_state.sbert_searched_df_p2)):
|
208 |
+
st.header(str(i+1)+'. '+st.session_state.sbert_searched_df_p2.iloc[i]['title'])
|
209 |
+
st.markdown('**Keywords :** '+ st.session_state.sbert_searched_df_p2.iloc[i]['vote_keywords'])
|
210 |
+
st.markdown(f"[Page source (Click here.)]({st.session_state.sbert_searched_df_p2.iloc[i].url})")
|
211 |
+
|
212 |
+
try:
|
213 |
+
banner_url = urllib.parse.quote(st.session_state.sbert_searched_df_p2.iloc[i]['banner'], safe=':/')
|
214 |
+
response = requests.get(banner_url,timeout=5)
|
215 |
+
st.image(response.content)
|
216 |
+
except:
|
217 |
+
st.image('https://icon-library.com/images/no-photo-icon/no-photo-icon-1.jpg')
|
218 |
+
finally:
|
219 |
+
st.write("---------------------------------------------------------------------------------------")
|
220 |
+
|
221 |
+
st.form_submit_button('Submit',on_click=set_state_p2,args=(0,))
|
222 |
+
|
223 |
+
#main
|
224 |
+
def main():
|
225 |
+
st.sidebar.title("Navigation")
|
226 |
+
page = st.sidebar.selectbox("Select a page:", ("Recommendation System", "Search Engine"))
|
227 |
+
|
228 |
+
if page == "Recommendation System":
|
229 |
+
st.session_state.current_page = 1
|
230 |
+
else:
|
231 |
+
st.session_state.current_page = 2
|
232 |
+
|
233 |
+
if page == "Recommendation System" and st.session_state.current_page == 1:
|
234 |
+
page1_recommendation()
|
235 |
+
elif page == "Search Engine" and st.session_state.current_page == 2:
|
236 |
+
page2_search_engine()
|
237 |
+
|
238 |
+
if __name__ == "__main__":
|
239 |
+
main()
|