Update app.py

app.py

@@ -6,11 +6,10 @@ from catboost import CatBoostRegressor
 def predict_mode():
     st.title("📊 Predict Mode: Salary Prediction")
     
-    # Input fields
+    # Input fields for prediction
     gender = st.selectbox("Gender", ["m", "f"])
     perc_10 = st.number_input("10th Percentage", min_value=0.0, max_value=100.0, value=80.0)
     perc_12 = st.number_input("12th Percentage", min_value=0.0, max_value=100.0, value=80.0)
-    
     tier = st.selectbox("College Tier (Only 1 or 2 accepted)", [1, 2])
     
     specializations = [
@@ -53,7 +52,7 @@ def predict_mode():
     ]
     location = st.selectbox("Location", locations)
     
-    # Prepare input DataFrame for prediction
+    # Prepare input DataFrame (ensure column names match training data)
     input_df = pd.DataFrame({
         "Gender": [gender],
         "10percentage": [perc_10],
@@ -73,10 +72,8 @@ def predict_mode():
 def db_mode():
     st.title("🗄️ DB Mode: Database Viewer")
     
-    # Connect to SQLite database.
+    # Connect to the SQLite database
     conn = sqlite3.connect("PlaceMeNot.db")
-    
-    # Get table names from the database
     table_query = "SELECT name FROM sqlite_master WHERE type='table';"
     tables = pd.read_sql(table_query, conn)
     
@@ -84,20 +81,50 @@ def db_mode():
         st.error("No tables found in the database.")
         return
     else:
-        # Assuming we use the first available table.
+        # Use first available table (you can extend this to allow user selection)
         table_name = tables.iloc[0, 0]
         st.write(f"Using table: **{table_name}**")
     
-    # Allow user to sort data
     cols = ["Gender", "10percentage", "12percentage", "CollegeTier", 
             "Specialization", "collegeGPA", "CollegeState", "Salary"]
+    
+    # Option to select which columns to display
+    selected_cols = st.multiselect("Select columns to display", options=cols, default=cols)
+    
+    # Sorting feature: note the need to quote columns with numeric names.
     sort_col = st.selectbox("Sort by", cols)
     order = st.radio("Order", ("Ascending", "Descending"))
     order_str = "ASC" if order == "Ascending" else "DESC"
     
-    query = f"SELECT * FROM {table_name} ORDER BY {sort_col} {order_str}"
-    df = pd.read_sql(query, conn)
-    st.dataframe(df)
+    # Filtering feature
+    st.subheader("Filtering")
+    filter_col = st.selectbox("Filter column", cols)
+    filter_value = st.text_input("Filter value (exact match, leave empty for no filter)")
+    
+    numeric_cols = ["10percentage", "12percentage", "CollegeTier", "collegeGPA", "Salary"]
+    filter_clause = ""
+    if filter_value:
+        if filter_col in numeric_cols:
+            filter_clause = f'WHERE "{filter_col}" = {filter_value}'
+        else:
+            filter_clause = f'WHERE "{filter_col}" = "{filter_value}"'
+    
+    # Build the query using double quotes for column names
+    col_str = ", ".join([f'"{col}"' for col in selected_cols])
+    query = f'SELECT {col_str} FROM "{table_name}" {filter_clause} ORDER BY "{sort_col}" {order_str}'
+    
+    st.write("Executing query:")
+    st.code(query, language="sql")
+    
+    try:
+        df = pd.read_sql(query, conn)
+        st.dataframe(df)
+        
+        st.subheader("Summary Statistics")
+        if not df.empty:
+            st.write(df.describe())
+    except Exception as e:
+        st.error(f"Error executing query: {e}")
     
     conn.close()