Move data loading up the tree

app.py

@@ -3,6 +3,7 @@ import page_home
 import page_likert
 import page_demographics
 from urllib.parse import quote, unquote
+from datasets import load_dataset
 
 # Default to wide mode
 st.set_page_config(layout="wide")
@@ -14,6 +15,15 @@ if 'page' not in st.session_state:
     initial_page = unquote(query_params.get("page", ["Home"])[0])
     st.session_state['page'] = initial_page
 
+
+@st.cache_data
+def load_data():
+    # Load data from Huggingface
+    dataset = load_dataset(
+        "krishaamer/taiwanese-college-students", data_files={'train': 'clean.csv'})
+    return dataset['train'].to_pandas()
+
+
 # Sidebar navigation using buttons
 st.sidebar.title("Navigation")
 if st.sidebar.button("Introduction"):
@@ -27,13 +37,15 @@ if st.sidebar.button("Student Attitudes"):
 st.experimental_set_query_params(page=quote(st.session_state['page']))
 
 # Page contents based on session state
+df = load_data()
+
 if st.session_state['page'] == 'Home':
-    page_home.show()
+    page_home.show(df)
 elif st.session_state['page'] == 'Likert':
-    page_likert.show()
+    page_likert.show(df)
 elif st.session_state['page'] == 'Demographics':
-    page_demographics.show()
+    page_demographics.show(df)
 
 # Rerun Calculations
 if st.sidebar.button("Rerun Calculations"):
-    st.rerun()
+    st.rerun()

page_demographics.py

@@ -2,14 +2,9 @@ import streamlit as st
 import pandas as pd
 from datasets import load_dataset
 
-@st.cache_data
-def show():
-    # Load data from Huggingface
-    dataset = load_dataset(
-        "krishaamer/taiwanese-college-students", data_files={'train': 'clean.csv'})
 
-    # Convert the loaded dataset to a pandas DataFrame
-    df = dataset['train'].to_pandas()
+@st.cache_data
+def show(df):
 
     # Generate and display the university ranking table
     generate_university_ranking_table(df)

page_likert.py

@@ -5,19 +5,14 @@ import seaborn as sns
 from datasets import load_dataset
 from matplotlib.font_manager import FontProperties
 
+
 @st.cache_data
-def show():
+def show(df):
 
     # Chinese font
     chinese_font = FontProperties(fname='mingliu.ttf')
 
-    # Get Data
-    dataset = load_dataset("krishaamer/taiwanese-college-students", data_files={'train': 'clean.csv'})
-
-    if dataset is not None:
-
-        # Load the CSV file into a DataFrame
-        df = dataset['train'].to_pandas()
+    if df is not None:
 
         likert_fields = {
             '購物習慣': [
@@ -146,11 +141,13 @@ def show():
 
         # Loop through each category in likert_fields to create visualizations
         for category, fields in likert_fields.items():
-            st.subheader(f'Distribution of Responses for {translation_mapping[category]}')
+            st.subheader(
+                f'Distribution of Responses for {translation_mapping[category]}')
 
             # Calculate the number of rows needed for this category
             num_fields = len(fields)
-            num_rows = -(-num_fields // 2)  # Equivalent to ceil(num_fields / 2)
+            # Equivalent to ceil(num_fields / 2)
+            num_rows = -(-num_fields // 2)
 
             # Create subplots with 2 columns for this category
             fig, axs = plt.subplots(num_rows, 2, figsize=(15, 5 * num_rows))
@@ -162,12 +159,14 @@ def show():
             # Loop through each field in the category to create individual bar plots
             for i, field in enumerate(fields):
                 # Create the bar plot
-                sns.countplot(x=f"{field} ({field_translation_mapping[category][i]})", data=df_translated, ax=axs[i])
+                sns.countplot(
+                    x=f"{field} ({field_translation_mapping[category][i]})", data=df_translated, ax=axs[i])
 
                 # Add title and labels
                 title_chinese = field
                 title_english = field_translation_mapping[category][i]
-                axs[i].set_title(f"{title_chinese}\n{title_english}", fontproperties=chinese_font)
+                axs[i].set_title(
+                    f"{title_chinese}\n{title_english}", fontproperties=chinese_font)
                 axs[i].set_xlabel('Likert Scale')
                 axs[i].set_ylabel('Frequency')
 
@@ -176,4 +175,4 @@ def show():
                 fig.delaxes(axs[i])
 
             # Show the plot
-            st.pyplot(fig)
+            st.pyplot(fig)