Update app.py

app.py

@@ -1,10 +1,24 @@
+# import os
+# import sys
+# from random import randint
+# import time
+# import uuid
+# import argparse
+# import streamlit as st
+# sys.path.append(os.path.abspath("../supv"))
+# from matumizi.util import *
+# from mcclf import *
+
 import os
 import sys
 from random import randint
 import time
 import uuid
 import argparse
+import pandas as pd
 import streamlit as st
+
+# Add the directory containing the required modules to sys.path
 sys.path.append(os.path.abspath("../supv"))
 from matumizi.util import *
 from mcclf import *
@@ -77,31 +91,72 @@ def  genVisitHistory(numUsers, convRate, label):
                 userSess.append(sessSummary)
 
         print(",".join(userSess))
+        
+# def trainModel(mlfpath):
+#     model = MarkovChainClassifier(mlfpath)
+#     model.train()
+
+# def predictModel(mlfpath):
+#     model = MarkovChainClassifier(mlfpath)
+#     model.predict()
+
 def trainModel(mlfpath):
     model = MarkovChainClassifier(mlfpath)
     model.train()
+    return model
+
 
-def predictModel(mlfpath):
+def predictModel(mlfpath, userID):
     model = MarkovChainClassifier(mlfpath)
-    model.predict()
-
-def main():
-    st.title("Markov Chain Classifier")
-
-    # Add input fields for command line arguments
-    op = st.selectbox("Operation", ["gen", "train", "pred"])
-    numUsers = st.slider("Number of Users", 1, 1000, 100)
-    convRate = st.slider("Conversion Rate", 1, 100, 10)
-    label = st.checkbox("Add Label")
-    mlfpath = st.text_input("ML Config File Path", value="false")
-
-    # Call functions based on selected operation
-    if op == "gen":
-        st.button("Generate Visit History", on_click=lambda: genVisitHistory(numUsers, convRate, label))
-    elif op == "train":
-        st.button("Train Model", on_click=lambda: trainModel(mlfpath))
-    elif op == "pred":
-        st.button("Predict Model", on_click=lambda: predictModel(mlfpath))
+    res = model.predict(userID)
+    return res
 
 if __name__ == "__main__":
-    main()
+    st.title("Conversion Prediction App")
+    st.write("Welcome to the Conversion Prediction App. This app uses a Markov chain based classifier to predict whether a customer will convert or not based on their visit history.")
+
+    op = st.sidebar.selectbox("Select Operation", ["Generate Visit History", "Train Model", "Predict"])
+
+    if op == "Generate Visit History":
+        st.write("Enter the parameters to generate the visit history:")
+        numUsers = st.number_input("Number of users", min_value=1, max_value=1000, value=100, step=1)
+        convRate = st.number_input("Conversion Rate (in percentage)", min_value=0, max_value=100, value=10, step=1)
+        label = st.checkbox("Add Labels")
+        st.write("Click the button below to generate the visit history")
+        if st.button("Generate"):
+            genVisitHistory(numUsers, convRate, label)
+            
+    elif op == "Train Model":
+        st.write("Train the model using the following parameters:")
+        mlfpath = st.text_input("MLF Path")
+        if st.button("Train"):
+            trainModel(mlfpath)
+
+    elif op == "Predict":
+        st.write("Predict using the trained model:")
+        mlfpath = st.text_input("MLF Path")
+        userID = st.text_input("User ID")
+        if st.button("Predict"):
+            result = predictModel(mlfpath, userID)
+            st.write("Prediction Result: ", result)
+
+# def main():
+#     st.title("Markov Chain Classifier")
+
+#     # Add input fields for command line arguments
+#     op = st.selectbox("Operation", ["gen", "train", "pred"])
+#     numUsers = st.slider("Number of Users", 1, 1000, 100)
+#     convRate = st.slider("Conversion Rate", 1, 100, 10)
+#     label = st.checkbox("Add Label")
+#     mlfpath = st.text_input("ML Config File Path", value="false")
+
+#     # Call functions based on selected operation
+#     if op == "gen":
+#         st.button("Generate Visit History", on_click=lambda: genVisitHistory(numUsers, convRate, label))
+#     elif op == "train":
+#         st.button("Train Model", on_click=lambda: trainModel(mlfpath))
+#     elif op == "pred":
+#         st.button("Predict Model", on_click=lambda: predictModel(mlfpath))
+
+# if __name__ == "__main__":
+#     main()