Update app.py

app.py

@@ -190,7 +190,6 @@ def ev_insights():
 
 
 def fuel_norm_distribution_dashboard(root_dir, state_mapping_file, start_year=2009, end_year=2025):
-
     # Folder paths
     folders = {
         "statewise": os.path.join(root_dir, "Fuel_vs_state"),
@@ -210,11 +209,11 @@ def fuel_norm_distribution_dashboard(root_dir, state_mapping_file, start_year=20
         file_path = os.path.join(folder_path, year_file)
         if os.path.exists(file_path):
             df = pd.read_csv(file_path)
-            # Remove 'Total' row/column if present
             if "Total" in df.columns:
                 df = df.drop(columns=["Total"])
             if df.iloc[:, 0].str.contains("Total", na=False).any():
                 df = df[~df.iloc[:, 0].str.contains("Total", na=False)]
+            df["Year"] = int(year_file.split(".")[0])
             return df
         else:
             st.warning(f"File not found: {file_path}")
@@ -229,107 +228,87 @@ def fuel_norm_distribution_dashboard(root_dir, state_mapping_file, start_year=20
             combined_df = pd.concat([combined_df, df], ignore_index=True)
         return combined_df
 
-    # Load data for selected year range
+    # Load all data
     statewise_df = load_data_for_year_range(folders["statewise"], start_year, end_year)
     norms_df = load_data_for_year_range(folders["norms"], start_year, end_year)
     fuelwise_df = load_data_for_year_range(folders["fuelwise"], start_year, end_year)
 
-    # Set up Streamlit app title
     st.title("🛢 Fuel Allocation & Emission Norms Dashboard (2009–2025)")
 
-    # Tabs for different visualizations
     tab1, tab2, tab3 = st.tabs([
         "📍 Fuel Distribution over States",
         "📊 Norm Distribution by State",
         "🔥 Emission by Fuel Type"
     ])
 
-    # Tab 1: Statewise Fuel Allocation
+    # Tab 1
     with tab1:
         st.subheader("Fuel Distribution Across States")
         if not statewise_df.empty:
-            # Convert from wide to long format
-            melted_df = pd.melt(statewise_df, id_vars=["Fuel"], var_name="State", value_name="Fuel_Amount")
-
-            # Replace state codes with full state names (case-insensitive matching)
+            melted_df = pd.melt(statewise_df, id_vars=["Fuel", "Year"], var_name="State", value_name="Fuel_Amount")
             melted_df["State"] = melted_df["State"].apply(lambda x: state_dict.get(x.upper(), x))
 
-            # Allow users to select states (categories)
-            selected_states = st.multiselect("Select States", ["Select All"] + list(melted_df["State"].unique()), default=[], key="state_selection")
-
-            # If "Select All" is chosen, select all states
+            selected_states = st.multiselect("Select States", ["Select All"] + sorted(melted_df["State"].unique()), default=[], key="state_selection")
             if "Select All" in selected_states:
                 selected_states = list(melted_df["State"].unique())
 
-            filtered_df = melted_df[melted_df["State"].isin(selected_states)]
+            year_range = st.slider("Select Year Range", min_value=start_year, max_value=end_year, value=(start_year, end_year), key="fuel_state_year")
+            filtered_df = melted_df[(melted_df["State"].isin(selected_states)) &
+                                    (melted_df["Year"].between(year_range[0], year_range[1]))]
 
-            # Create and display the plot
             fig1 = px.bar(filtered_df, x="State", y="Fuel_Amount", color="Fuel",
-                          title=f"Fuel Distribution from {start_year} to {end_year}",
+                          title=f"Fuel Distribution from {year_range[0]} to {year_range[1]}",
                           labels={"Fuel_Amount": "Amount (in units)"}, barmode="group")
             st.plotly_chart(fig1, use_container_width=True)
 
-    # Tab 2: Norm Distribution by State
+    # Tab 2
     with tab2:
         st.subheader("Emission Norm Distribution by State")
         if not norms_df.empty:
-            # Convert from wide to long format
-            melted_norms = pd.melt(norms_df, id_vars=["Norms"], var_name="State", value_name="Count")
-
-            # Replace state codes with full state names (case-insensitive matching)
+            melted_norms = pd.melt(norms_df, id_vars=["Norms", "Year"], var_name="State", value_name="Count")
             melted_norms["State"] = melted_norms["State"].apply(lambda x: state_dict.get(x.upper(), x))
 
-            # Allow users to select a state
-            selected_states = st.multiselect("Select States", ["Select All"] + list(melted_norms["State"].unique()), default=[], key="norm_state_selection")
-
-            # If "Select All" is chosen, select all states
+            selected_states = st.multiselect("Select States", ["Select All"] + sorted(melted_norms["State"].unique()), default=[], key="norm_state_selection")
             if "Select All" in selected_states:
                 selected_states = list(melted_norms["State"].unique())
 
-            filtered_norms = melted_norms[melted_norms["State"].isin(selected_states)]
+            year_range = st.slider("Select Year Range", min_value=start_year, max_value=end_year, value=(start_year, end_year), key="norm_state_year")
+            filtered_norms = melted_norms[(melted_norms["State"].isin(selected_states)) &
+                                          (melted_norms["Year"].between(year_range[0], year_range[1]))]
 
-            # Create and display the plot
             fig2 = px.bar(filtered_norms, x="State", y="Count", color="Norms",
-                          title=f"Emission Norms Distribution from {start_year} to {end_year}",
+                          title=f"Emission Norms Distribution from {year_range[0]} to {year_range[1]}",
                           labels={"Count": "Count of Norms"}, barmode="group")
             st.plotly_chart(fig2, use_container_width=True)
 
-    # Tab 3: Fuelwise Emissions
+    # Tab 3
     with tab3:
         st.subheader("Norm Emissions by Fuel Type")
         if not fuelwise_df.empty:
-            # Check if 'Fuel' exists and others are norms
-            norm_columns = [col for col in fuelwise_df.columns if col != "Fuel"]
-            if "Fuel" in fuelwise_df.columns and len(norm_columns) > 0:
-                # Melt the DataFrame to long format
-                melted_fuelwise = pd.melt(fuelwise_df, id_vars=["Fuel"], 
+            norm_columns = [col for col in fuelwise_df.columns if col not in ["Fuel", "Year"]]
+            if "Fuel" in fuelwise_df.columns and norm_columns:
+                melted_fuelwise = pd.melt(fuelwise_df, id_vars=["Fuel", "Year"],
                                           var_name="Norm_Type", value_name="Emission")
-
-                # Drop missing or non-numeric values if necessary
                 melted_fuelwise = melted_fuelwise.dropna()
                 melted_fuelwise["Emission"] = pd.to_numeric(melted_fuelwise["Emission"], errors="coerce")
                 melted_fuelwise = melted_fuelwise.dropna(subset=["Emission"])
 
-                # Allow users to select fuel categories
-                selected_fuels = st.multiselect("Select Fuel Types", ["Select All"] + list(melted_fuelwise["Fuel"].unique()), default=[], key="fuel_selection")
-
-                # If "Select All" is chosen, select all fuels
+                selected_fuels = st.multiselect("Select Fuel Types", ["Select All"] + sorted(melted_fuelwise["Fuel"].unique()), default=[], key="fuel_selection")
                 if "Select All" in selected_fuels:
                     selected_fuels = list(melted_fuelwise["Fuel"].unique())
 
-                filtered_fuelwise = melted_fuelwise[melted_fuelwise["Fuel"].isin(selected_fuels)]
+                year_range = st.slider("Select Year Range", min_value=start_year, max_value=end_year, value=(start_year, end_year), key="fuel_year")
+                filtered_fuelwise = melted_fuelwise[(melted_fuelwise["Fuel"].isin(selected_fuels)) &
+                                                    (melted_fuelwise["Year"].between(year_range[0], year_range[1]))]
 
-                # Create and display the plot
                 fig3 = px.bar(filtered_fuelwise, x="Fuel", y="Emission", color="Norm_Type",
-                              title=f"Emission per Fuel Type from {start_year} to {end_year}",
+                              title=f"Emission per Fuel Type from {year_range[0]} to {year_range[1]}",
                               labels={"Emission": "Emission Amount"}, barmode="stack")
                 st.plotly_chart(fig3, use_container_width=True)
 
-                # Additional visualization for Top Fuels emitting most pollution
-                top_fuels = melted_fuelwise.groupby("Fuel")["Emission"].sum().reset_index()
-                top_fuels_sorted = top_fuels.sort_values(by="Emission", ascending=False).head(10)
-
                 st.subheader("Top Fuels Emitting Most Pollution")
+                top_fuels = filtered_fuelwise.groupby("Fuel")["Emission"].sum().reset_index()
+                top_fuels_sorted = top_fuels.sort_values(by="Emission", ascending=False).head(10)
                 st.write(top_fuels_sorted)
 
                 fig4 = px.bar(top_fuels_sorted, x="Fuel", y="Emission", title="Top Fuels by Emission",