new-app-2 (#4)

- New app with lesser features on interface (a83ef1896eeacde625fb227e4356a860d011c287)
- Button click calls api (f3dd5c4a7cc7a197a48e1d99c6ffbf662ab852de)


Co-authored-by: Bright Eshun <[email protected]>

app.py

@@ -1,69 +1,57 @@
 # Loading key libraries
 import streamlit as st
 import os
-import pickle
 import numpy as np
 import pandas as pd
-import re
-from pathlib import Path
+
 from PIL import Image
 import matplotlib.pyplot as plt
 import seaborn as sns
 import requests
-
-
+import datetime 
 
 
 # set api endpoint
-URL = 'https://bright1-sales-forecasting-api.hf.space'
+URL = 'https://bright1-sales-forecasting-ap1-2.hf.space'
 API_ENDPOINT = '/predict'
 
+# get list/choices for inputs 
+CITIES = ['Accra', 'Aflao', 'Akim Oda', 'Akwatia', 'Bekwai', 'Cape coast', 'Elmina,', 'Gbawe', 'Ho', 'Hohoe', 'intampo', 'Koforidua', 'Kumasi', 'Mampong', 'Obuasi', 'Prestea', 'Suhum', 'Tamale', 'Techiman', 'Tema', 'Teshie', 'Winneba']
+CLUSTER = [ i for i in range(0, 17)]
+STORE_ID = [ i for i in range(1, 55)]
+CATEGORY_ID = [ i for i in range(0, 35)]
+
+
 # Setting the page configurations
 st.set_page_config(page_title = "Prediction Forecasting", layout= "wide", initial_sidebar_state= "auto")
 
 # Setting the page title
 st.title("Grocery Store Forecasting Prediction")
 
-# Load the saved data
-df = pd.read_csv('Grocery.csv')
-
 
+# src\app\images1.jpg
 image1 = Image.open('images1.jpg')
-image2 = Image.open('image 2.jpg')
 
-def make_prediction(store_id, category_id, onpromotion, year,month, dayofmonth, 
-                    dayofweek, dayofyear,weekofyear, quarter, is_month_start, is_month_end,
-                    is_quarter_start, is_quarter_end, is_year_start, is_year_end, 
-                    year_weekofyear,city, store_type, cluster):
+
+def make_prediction(store_id, category_id, onpromotion, city, store_type, cluster, date):
     
     
     parameters = {
     'store_id':int(store_id), 
     'category_id':int(category_id), 
     'onpromotion' :int(onpromotion),
-    'year' : int(year), 
-    'month' : int(month), 
-    'dayofmonth' :int(dayofmonth),
-    'dayofweek' : int(dayofweek),
-    'dayofyear' : int(dayofyear), 
-    'weekofyear' : int(weekofyear), 
-    'quarter' : int(quarter),
-    'is_month_start' : int(is_month_start),
-    'is_month_end' : int(is_month_end), 
-    'is_quarter_start' : int(is_quarter_start), 
-    'is_quarter_end' : int(is_quarter_end), 
-    'is_year_start' : int(is_year_start),
-    'is_year_end' : (is_year_end), 
-    'year_weekofyear' : int(year_weekofyear),
     'city' : city,
     'store_type' : int(store_type), 
     'cluster': int(cluster),
+    'date_': date,
 
     } 
 
-
+    # make a request to the api
     response = requests.post(url=f'{URL}{API_ENDPOINT}', params=parameters)
+    
     sales_value = response.json()['sales']
+
     sales_value = round(sales_value, 4)
     return sales_value
 
@@ -72,37 +60,22 @@ st.image(image1, width = 700)
 
 st.sidebar.markdown('User Input Details and Information')
 
-store_id= st.sidebar.selectbox('store_id', options = sorted(list(df['store_id'].unique())))
-category_id= st.sidebar.selectbox('categegory_id',options = sorted(list(df['category_id'].unique())))
-onpromotion= st.sidebar.number_input('onpromotion', min_value= df["onpromotion"].min(), value= df["onpromotion"].min())
-year = st.sidebar.selectbox('year', options = sorted(list(df['year'].unique())))
-month = st.sidebar.selectbox('month', options = sorted(list(df['month'].unique())))
-dayofmonth= st.sidebar.number_input('dayofmonth', min_value= df["dayofmonth"].min(), value= df["dayofmonth"].min())
-dayofweek = st.sidebar.number_input('dayofweek', min_value= df["dayofweek"].min(), value= df["dayofweek"].min())
-dayofyear = st.sidebar.number_input('dayofyear', min_value= df["dayofyear"].min(), value= df["dayofyear"].min())
-weekofyear = st.sidebar.number_input('weekofyear', min_value= df["weekofyear"].min(), value= df["weekofyear"].min())
-quarter  = st.sidebar.number_input('quarter', min_value= df["quarter"].min(), value= df["quarter"].min())
-is_month_start = st.sidebar.number_input('is_month_start', min_value= df["is_month_start"].min(), value= df["is_month_start"].min())
-is_month_end = st.sidebar.number_input('is_month_end', min_value= df["is_month_end"].min(), value= df["is_month_end"].min())
-is_quarter_start = st.sidebar.number_input('is_quarter_start', min_value= df["is_quarter_start"].min(), value= df["is_quarter_start"].min())
-is_quarter_end = st.sidebar.number_input('is_quarter_end', min_value= df["is_quarter_end"].min(), value= df["is_quarter_end"].min())
-is_year_start = st.sidebar.number_input('is_year_start', min_value= df["is_year_start"].min(), value= df["is_year_start"].min())
-is_year_end = st.sidebar.number_input('is_year_end', min_value= df["is_year_end"].min(), value= df["is_year_end"].min())
-year_weekofyear = st.sidebar.number_input('year_weekofyear', min_value= df["year_weekofyear"].min(), value= df["year_weekofyear"].min())
-city =  st.sidebar.selectbox("city:", options= sorted(set(df["city"])))
-store_type=  st.sidebar.number_input('type', min_value= df["type"].min(), value= df["type"].min())
-cluster = st.sidebar.selectbox('cluster', options = sorted(list(df['cluster'].unique())))
-
-
-
-# make prediction 
-sales_value = make_prediction(store_id, category_id, onpromotion, year,month, dayofmonth, 
-                  dayofweek, dayofyear,weekofyear, quarter, is_month_start, is_month_end, 
-                  is_quarter_start, is_quarter_end, is_year_start, is_year_end, 
-                  year_weekofyear,city, store_type, cluster)
+# Create interface
+date= st.sidebar.date_input("Enter the Date",datetime.date(2023, 6, 30))
+store_id= st.sidebar.selectbox('Store id', options=STORE_ID)
+category_id= st.sidebar.selectbox('categegory_id', options=CATEGORY_ID)
+onpromotion= st.sidebar.number_input('onpromotion', step=1)
+city =  st.sidebar.selectbox("city:", options= CITIES)
+store_type=  st.sidebar.selectbox('type', options=[0, 1, 2, 3, 4])
+cluster = st.sidebar.selectbox('cluster', options = CLUSTER )
+
+
+
 
 # get predicted value
-if st.button('Predict'):
+if st.sidebar.button('Predict', use_container_width=True, type='primary'):
+            # make prediction 
+            sales_value = make_prediction(store_id, category_id, onpromotion,city, store_type, cluster, date)
             st.success('The predicted target is ' + str(sales_value))