Update app.py

app.py

@@ -1,31 +1,30 @@
-import streamlit as st
 import pandas as pd
-import matplotlib.pyplot as plt
-import seaborn as sns
+import streamlit as st
+
+st.title('Udemy Courses Analysis')
+
+courses_info = pd.read_csv('udemy_courses_info.csv')
+courses_year = pd.read_csv('udemy_courses_year.csv')
 
-st.title('My first app')
-st.write("Here's our first app with Streamlit")
+couses_df = pd.merge(courses_info, courses_year, on='course_id')
 
-df_courses_year = pd.read_csv('udemy_courses_year.csv')
+options = st.multiselect(
+    "Please select subjects",
+    couses_df.subject.unique(),
+    couses_df.subject.unique())
 
-# Extract unique years and sort them
-years = sorted(df_courses_year['year'].unique())
+couses_df = couses_df[couses_df['subject'].isin(options)]
 
-# Sidebar for selecting year(s)
-selected_years = st.sidebar.multiselect('Select Year(s)', years, default=years)
+# Group the DataFrame by 'subject' and 'course_title', and calculate the sum of 'num_subscribers'
+grouped = couses_df.groupby(['subject', 'course_title'])['num_subscribers'].sum()
 
-# Filter data based on selected year(s)
-filtered_data = df_courses_year[df_courses_year['year'].isin(selected_years)]
+# Reset the index of the grouped DataFrame
+grouped = grouped.reset_index()
 
-# Set style for better visuals
-sns.set(style="whitegrid")
+# Sort the grouped DataFrame by 'subject' and 'num_subscribers' in descending order
+grouped = grouped.sort_values(['subject', 'num_subscribers'], ascending=[True, False])
 
-# Create the plot
-plt.figure(figsize=(10, 6))
-ax = sns.countplot(x='year', data=filtered_data, palette='viridis')
-ax.set_title('Count of Observations per Year')
-ax.set_xlabel('Year')
-ax.set_ylabel('Count')
-plt.xticks(rotation=45)
+# Group the sorted DataFrame by 'subject' and get the top 5 rows for each group
+top_courses = grouped.groupby('subject').head(5)
 
-st.pyplot(plt)
+st.dataframe(top_courses)