Upload eda.py

eda.py

@@ -36,27 +36,6 @@ def app():
 
     object_columns = df.select_dtypes(include=['object']).columns
     numerical_columns = df.select_dtypes(exclude=['object']).columns
-    
-    #plot all object columns to multiple boxplots based on index result of object_columns
-    for i in range(0,len(object_columns),2):
-        if len(object_columns) > i+1:
-            fig = plt.figure(figsize=(10,4))
-            plt.subplot(121)
-            sns.countplot(df[object_columns[i]])
-            plt.xticks(rotation=90)
-            plt.subplot(122)
-            sns.countplot(df[object_columns[i+1]])
-            plt.xticks(rotation=90)
-            plt.tight_layout()
-            # plt.show()
-            st.pyplot(fig)
-
-        else:
-            sns.countplot(df[object_columns[i]])
-            plt.xticks(rotation=90)
-            plt.show()
-            st.pyplot(fig)
-
 
     # membuat barplot
     # st.write('#### Plot AttackingWorkRate')
@@ -64,29 +43,40 @@ def app():
     # sns.countplot(x='AttackingWorkRate', data=df)
     # st.pyplot(fig)
 
-    # # membuat hist
-    # st.write('#### Histogram of Rating')
-    # fig = plt.figure(figsize=(15,5))
-    # sns.histplot(df['Overall'], bins=30, kde=True)
-    # st.pyplot(fig)
 
     # # Membuat hist berdasarkan input
     # st.write('#### Select histogram input')
 
-    # option 2 (radiobutton) selectbox -> radio
-    # limit_balance                
-    # sex                         
-    # education_level               
-    # martial_status              
-    # age 
-    option = st.selectbox('Select Column:', ('limit_balance', 'sex', 'education_level', 'martial_status', 'age'))
+    st.write('#### Plot Categorical Columns')
+    option_cat = st.selectbox('Select Column:', ('sex', 'education_level', 'martial_status'))
+    fig = plt.figure(figsize=(15,5))
+    sns.countplot(x=option_cat, data=df)
+    st.pyplot(fig)
+    
+    st.write('#### Plot Numerical Columns')
+    option = st.selectbox('Select Column:', ('limit_balance', 'age'))
     fig = plt.figure(figsize=(15,5))
     sns.histplot(df[option], bins=30, kde=True)
     st.pyplot(fig)
+    
+    option_pay = st.selectbox('Select Column:', ('pay_0', 'pay_2', 'pay_3', 'pay_4', 'pay_5', 'pay_6'))
+    fig = plt.figure(figsize=(15,5))
+    sns.histplot(df[option_pay], bins=30, kde=True)
+    st.pyplot(fig)
+    
+    option_bill_amt = st.selectbox('Select Column:', ('bill_amt_1', 'bill_amt_2', 'bill_amt_3', 'bill_amt_4', 'bill_amt_5', 'bill_amt_6'))
+    fig = plt.figure(figsize=(15,5))
+    sns.histplot(df[option_bill_amt], bins=30, kde=True)
+    st.pyplot(fig)
+
+    option_pay_amt = st.selectbox('Select Column:', ('pay_amt_1', 'pay_amt_2', 'pay_amt_3', 'pay_amt_4', 'pay_amt_5', 'pay_amt_6'))
+    fig = plt.figure(figsize=(15,5))
+    sns.histplot(df[option_bill_amt], bins=30, kde=True)
+    st.pyplot(fig)
 
     # Membuat plotly plot
-    st.write('#### Plotly Plot - Education Level vs default_payment_next_month')
-    fig  = px.scatter(df, x = 'education_level', y = 'default_payment_next_month', hover_data = ['sex', 'age'])
+    st.write('#### Plotly Plot - Age vs Limit Balance')
+    fig  = px.scatter(df, x = 'age', y = 'limit_balance', hover_data = ['sex', 'default_payment_next_month'])
     st.plotly_chart(fig)
 
 if __name__ == '__main__':