Update app.py

app.py

@@ -3,75 +3,66 @@ import pandas as pd
 import plotly.graph_objects as go
 import matplotlib.pyplot as plt
 import seaborn as sns
+import os
 
 st.set_page_config(page_title="Legislative Visualizations", layout="wide")
 
 st.title("Legislative Bill Analysis Dashboard")
 
-st.sidebar.markdown("### ⚙️ Demo Mode")
-use_sample = st.sidebar.checkbox("Use sample dataset")
+DATA_PATH = "Illinois_Entire_Data_Insights_Final_v2.csv"
 
-if use_sample:
-    st.info("Using embedded sample dataset for demo.")
-    df = pd.DataFrame({
-        'intent_standardized': ['Fund', 'Fund', 'Regulate', 'Promote', 'Regulate'],
-        'stance_standardized': ['Support', 'Neutral', 'Oppose', 'Support', 'Oppose'],
-        'intended_beneficiaries_standardized': ['Students', 'Schools', 'Teachers', 'Parents', 'Students'],
-        'category_&_subcategory_standardized': ['Education > Higher Ed'] * 5,
-        'policy_impact_areas_standardized': ['Access', 'Infrastructure', 'Curriculum', 'Affordability', 'Access']
-    })
-else:
-    uploaded_file = st.file_uploader("Upload legislative dataset (.csv)", type=["csv"])
-    if uploaded_file:
-        df = pd.read_csv(uploaded_file)
-    else:
-        st.info("Please upload a dataset file to view the visualizations.")
-        st.stop()
+if os.path.exists(DATA_PATH):
+    df = pd.read_csv(DATA_PATH)
+    st.success(f"Loaded dataset from {DATA_PATH}")
 
-# Sankey Diagram
-st.header("🔗 Sankey Diagram: Intent → Stance → Beneficiaries")
-sankey_df = df[['intent_standardized', 'stance_standardized', 'intended_beneficiaries_standardized']].dropna()
+    # Sankey Diagram
+    st.header("🔗 Sankey Diagram: Intent → Stance → Beneficiaries")
+    sankey_df = df[['intent_standardized', 'stance_standardized', 'intended_beneficiaries_standardized']].dropna()
 
-if not sankey_df.empty:
-    labels = list(pd.unique(sankey_df['intent_standardized'].tolist() +
-                            sankey_df['stance_standardized'].tolist() +
-                            sankey_df['intended_beneficiaries_standardized'].tolist()))
-    label_map = {label: i for i, label in enumerate(labels)}
+    if not sankey_df.empty:
+        labels = list(pd.unique(sankey_df['intent_standardized'].tolist() +
+                                sankey_df['stance_standardized'].tolist() +
+                                sankey_df['intended_beneficiaries_standardized'].tolist()))
+        label_map = {label: i for i, label in enumerate(labels)}
 
-    intent_stance = sankey_df.groupby(['intent_standardized', 'stance_standardized']).size().reset_index(name='count')
-    stance_beneficiary = sankey_df.groupby(['stance_standardized', 'intended_beneficiaries_standardized']).size().reset_index(name='count')
+        intent_stance = sankey_df.groupby(['intent_standardized', 'stance_standardized']).size().reset_index(name='count')
+        stance_beneficiary = sankey_df.groupby(['stance_standardized', 'intended_beneficiaries_standardized']).size().reset_index(name='count')
 
-    source = intent_stance['intent_standardized'].map(label_map).tolist() + stance_beneficiary['stance_standardized'].map(label_map).tolist()
-    target = intent_stance['stance_standardized'].map(label_map).tolist() + stance_beneficiary['intended_beneficiaries_standardized'].map(label_map).tolist()
-    value  = intent_stance['count'].tolist() + stance_beneficiary['count'].tolist()
+        source = intent_stance['intent_standardized'].map(label_map).tolist() + stance_beneficiary['stance_standardized'].map(label_map).tolist()
+        target = intent_stance['stance_standardized'].map(label_map).tolist() + stance_beneficiary['intended_beneficiaries_standardized'].map(label_map).tolist()
+        value  = intent_stance['count'].tolist() + stance_beneficiary['count'].tolist()
 
-    fig_sankey = go.Figure(data=[go.Sankey(
-        node=dict(pad=15, thickness=20, line=dict(color="black", width=0.5), label=labels),
-        link=dict(source=source, target=target, value=value)
-    )])
-    fig_sankey.update_layout(title_text="Sankey: Intent → Stance → Beneficiary", font_size=12)
+        fig_sankey = go.Figure(data=[go.Sankey(
+            node=dict(pad=15, thickness=20, line=dict(color="black", width=0.5), label=labels),
+            link=dict(source=source, target=target, value=value)
+        )])
+        fig_sankey.update_layout(title_text="Sankey: Intent → Stance → Beneficiary", font_size=12)
 
-    st.plotly_chart(fig_sankey, use_container_width=True)
-else:
-    st.warning("Sankey input columns contain only null values or are missing.")
+        st.plotly_chart(fig_sankey, use_container_width=True)
+    else:
+        st.warning("Sankey input columns contain only null values or are missing.")
+
+    # Heatmap
+    st.header("🧯 Heatmap: Category vs Policy Impact Area")
+    heat_df = df[['category_&_subcategory_standardized', 'policy_impact_areas_standardized']].dropna()
 
-# Heatmap
-st.header("🧯 Heatmap: Category vs Policy Impact Area")
-heat_df = df[['category_&_subcategory_standardized', 'policy_impact_areas_standardized']].dropna()
+    if not heat_df.empty:
+        heat = heat_df.pivot_table(index='category_&_subcategory_standardized',
+                                   columns='policy_impact_areas_standardized',
+                                   aggfunc=len,
+                                   fill_value=0)
 
-if not heat_df.empty:
-    heat = heat_df.pivot_table(index='category_&_subcategory_standardized',
-                               columns='policy_impact_areas_standardized',
-                               aggfunc=len,
-                               fill_value=0)
+        plt.figure(figsize=(14, 8))
+        sns.heatmap(heat, cmap='coolwarm', annot=False)
+        plt.title("Heatmap: Category vs Policy Impact Area")
+        plt.xlabel("Policy Impact Area")
+        plt.ylabel("Category")
+        plt.tight_layout()
 
-    plt.figure(figsize=(14, 8))
-    sns.heatmap(heat, cmap='coolwarm', annot=False)
-    plt.title("Heatmap: Category vs Policy Impact Area")
-    plt.xlabel("Policy Impact Area")
-    plt.ylabel("Category")
-    plt.tight_layout()
+        st.pyplot(plt)
+    else:
+        st.warning("Heatmap input columns contain only null values or are missing.")
 
-    st.pyplot(plt)
 else:
-    st.warning("Heatmap input columns contain only null values or are missing.")
+    st.error(f"Data file {DATA_PATH} not found in Space repo.")
+    st.stop()