Update pages/2_Data_CLeaning_and_Preprocessing.py

pages/2_Data_CLeaning_and_Preprocessing.py

@@ -1,149 +1,134 @@
 import streamlit as st
 import pandas as pd
-import seaborn as sns
-import matplotlib.pyplot as plt
+import plotly.graph_objects as go
+import plotly.express as px
+from plotly.subplots import make_subplots
 
 # Page Title
 st.title("Exploratory Data Analysis (EDA) App")
 
 st.markdown("""
 ### Perform EDA and Clean Data
-Upload a CSV file to begin. This app will provide basic insights into the dataset, 
-highlight missing values, and visualize numeric and categorical columns.
+This app provides basic insights into the dataset, highlights missing values, 
+and visualizes numeric and categorical columns.
 ---
 """)
 
-# File Upload Section
-st.header("Upload Dataset")
-
-uploaded_file = st.file_uploader("Upload CSV", type=["csv"])
-
-# Check if file is uploaded
-if uploaded_file is not None:
-    if uploaded_file.size > 0:
-        try:
-            # Read the CSV file
-            data = pd.read_csv(uploaded_file)
-            st.session_state['df'] = data  # Store the data for use in other pages
-            st.success("Dataset uploaded successfully!")
-
-            # Show Data Preview
-            st.write("### Preview of Dataset")
-            st.dataframe(data.head())
-
-            # Overview Section
-            st.write("### Dataset Overview")
-            st.write(data.describe())
-
-            # Missing Values
-            st.write("### Missing Values")
-            st.write(data.isnull().sum())
-
-            # Duplicate Rows
-            st.write("### Duplicate Rows")
-            st.write(f"Number of duplicate rows: {data.duplicated().sum()}")
-
-            # Visualize Numeric Data
-            numeric_columns = data.select_dtypes(include=['float64', 'int64']).columns
-            if len(numeric_columns) > 0:
-                st.write("### Histograms for Numeric Columns")
-                for col in numeric_columns:
-                    fig, ax = plt.subplots()
-                    sns.histplot(data[col], kde=True, palette="crest", ax=ax)
-                    ax.set_title(f'Histogram of {col}')
-                    st.pyplot(fig)
-
-                st.write("### Boxplots for Numeric Columns")
-                for col in numeric_columns:
-                    fig, ax = plt.subplots()
-                    sns.boxplot(x=data[col], palette="mako", ax=ax)
-                    ax.set_title(f'Boxplot of {col}')
-                    st.pyplot(fig)
-            else:
-                st.write("No numeric columns available for visualization.")
-
-            # Visualize Categorical Data
-            categorical_columns = data.select_dtypes(include=['object', 'category']).columns
-            if len(categorical_columns) > 0:
-                st.write("### Bar Plots for Categorical Columns")
-                selected_cat_col = st.selectbox("Select a Categorical Column", categorical_columns)
-                
-                st.write(f"Value Counts for '{selected_cat_col}':")
-                st.write(data[selected_cat_col].value_counts())
-                
-                fig, ax = plt.subplots()
-                sns.countplot(x=selected_cat_col, data=data, palette="viridis", ax=ax)
-                ax.set_title(f'Bar Plot of {selected_cat_col}')
-                st.pyplot(fig)
-            else:
-                st.write("No categorical columns available for visualization.")
-
-            # Correlation Matrix
-            if len(numeric_columns) > 1:
-                st.write("### Correlation Matrix")
-                corr_matrix = data[numeric_columns].corr()
-                st.write(corr_matrix)
-
-                fig, ax = plt.subplots(figsize=(10, 8))
-                sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', ax=ax)
-                st.pyplot(fig)
-
-            # Check the columns before renaming
-            st.write("### Dataset Columns:")
-            st.write(data.columns)
-
-            # Renaming columns if they exist
-            if 'ProductCategory' in data.columns and 'ProductBrand' in data.columns and 'ProductPrice' in data.columns:
-                data = data.rename(columns={'ProductCategory': 'Category', 'ProductBrand': 'Brand', 'ProductPrice': 'Price'})
-                st.success("Columns renamed successfully!")
-            else:
-                st.warning("Columns 'ProductCategory', 'ProductBrand', or 'ProductPrice' not found in the dataset.")
-
-            # Now check if 'Category' exists and plot
-            if 'Category' in data.columns:
-                st.write("### Bar Plot for Category")
-                fig, ax = plt.subplots()
-                sns.countplot(x='Category', data=data, palette='plasma', ax=ax)
-                st.pyplot(fig)
-            else:
-                st.warning("'Category' column not found for plotting.")
-
-            # Binning of age column
-            bins = [0, 18, 35, 50, 65, 100]
-            labels = ['Child', 'Young Adult', 'Adult', 'Middle Aged', 'Senior']
-            data['age_bins'] = pd.cut(data['CustomerAge'], bins=bins, labels=labels, right=False)
-
-            # Data Cleaning Section
-            st.write("### Cleaned Dataset")
-            cleaned_data = data.drop_duplicates()
-            st.dataframe(cleaned_data)
-
-            # Save cleaned data to CSV and provide download option
-            cleaned_csv = cleaned_data.to_csv(index=False).encode('utf-8')
-            st.download_button(
-                label="Download Cleaned Dataset",
-                data=cleaned_csv,
-                file_name="cleaned_dataset.csv",
-                mime="text/csv"
+# Check if the dataset is already in session state
+if 'df' in st.session_state:
+    data = st.session_state['df']
+    st.success("Dataset loaded from previous session!")
+
+    # Show Data Preview
+    st.write("### Preview of Dataset")
+    st.dataframe(data.head())
+
+    # Overview Section
+    st.write("### Dataset Overview")
+    st.write(data.describe())
+
+    # Missing Values
+    st.write("### Missing Values")
+    st.write(data.isnull().sum())
+
+    # Duplicate Rows
+    st.write("### Duplicate Rows")
+    st.write(f"Number of duplicate rows: {data.duplicated().sum()}")
+
+    # Visualize Numeric Data
+    numeric_columns = data.select_dtypes(include=['float64', 'int64']).columns
+    if len(numeric_columns) > 0:
+        st.write("### Histograms for Numeric Columns")
+        
+        # Create subplots for histograms
+        fig = make_subplots(rows=len(numeric_columns), cols=1, subplot_titles=numeric_columns)
+        for i, col in enumerate(numeric_columns):
+            hist = px.histogram(data, x=col, nbins=30, title=f'Histogram of {col}')
+            fig.add_trace(
+                hist.data[0],
+                row=i+1, col=1
             )
 
-            # Store the cleaned dataframe in session state for use in other pages
-            st.session_state['df'] = cleaned_data
+        fig.update_layout(height=500 * len(numeric_columns), title_text="Histograms for Numeric Columns")
+        st.plotly_chart(fig)
+
+        st.write("### Boxplots for Numeric Columns")
+
+        # Create subplots for boxplots
+        fig = make_subplots(rows=len(numeric_columns), cols=1, subplot_titles=numeric_columns)
+        for i, col in enumerate(numeric_columns):
+            boxplot = px.box(data, y=col, title=f'Boxplot of {col}')
+            fig.add_trace(
+                boxplot.data[0],
+                row=i+1, col=1
+            )
 
-        except pd.errors.EmptyDataError:
-            st.error("The uploaded CSV file is empty. Please upload a valid file.")
-        except pd.errors.ParserError:
-            st.error("The file is not properly formatted as a CSV. Please check the data.")
-        except Exception as e:
-            st.error(f"An unexpected error occurred: {e}")
+        fig.update_layout(height=500 * len(numeric_columns), title_text="Boxplots for Numeric Columns")
+        st.plotly_chart(fig)
     else:
-        st.error("The uploaded file is empty.")
-else:
-    st.info("Upload a CSV file to get started.")
+        st.write("No numeric columns available for visualization.")
+
+    # Visualize Categorical Data
+    categorical_columns = data.select_dtypes(include=['object', 'category']).columns
+    if len(categorical_columns) > 0:
+        st.write("### Bar Plots for Categorical Columns")
+        selected_cat_col = st.selectbox("Select a Categorical Column", categorical_columns)
+
+        st.write(f"Value Counts for '{selected_cat_col}':")
+        st.write(data[selected_cat_col].value_counts())
+
+        fig = px.bar(data, x=selected_cat_col, title=f'Bar Plot of {selected_cat_col}', color=selected_cat_col)
+        st.plotly_chart(fig)
+    else:
+        st.write("No categorical columns available for visualization.")
+
+    # Correlation Matrix for Numeric Columns
+    if len(numeric_columns) > 1:
+        st.write("### Correlation Matrix")
+        corr_matrix = data[numeric_columns].corr()
+        fig = px.imshow(corr_matrix, title="Correlation Matrix", color_continuous_scale='coolwarm')
+        st.plotly_chart(fig)
+
+    # Check the columns before renaming
+    st.write("### Dataset Columns:")
+    st.write(data.columns)
+
+    # Renaming columns if they exist
+    if 'ProductCategory' in data.columns and 'ProductBrand' in data.columns and 'ProductPrice' in data.columns:
+        data = data.rename(columns={'ProductCategory': 'Category', 'ProductBrand': 'Brand', 'ProductPrice': 'Price'})
+        st.success("Columns renamed successfully!")
+    else:
+        st.warning("Columns 'ProductCategory', 'ProductBrand', or 'ProductPrice' not found in the dataset.")
+
+    # Now check if 'Category' exists and plot
+    if 'Category' in data.columns:
+        st.write("### Bar Plot for Category")
+        fig = px.bar(data, x='Category', title='Bar Plot of Category', color='Category')
+        st.plotly_chart(fig)
+    else:
+        st.warning("'Category' column not found for plotting.")
+
+    # Binning of age column
+    bins = [0, 18, 35, 50, 65, 100]
+    labels = ['Child', 'Young Adult', 'Adult', 'Middle Aged', 'Senior']
+    data['age_bins'] = pd.cut(data['CustomerAge'], bins=bins, labels=labels, right=False)
+
+    # Data Cleaning Section
+    st.write("### Cleaned Dataset")
+    cleaned_data = data.drop_duplicates()
+    st.dataframe(cleaned_data)
+
+    # Save cleaned data to CSV and provide download option
+    cleaned_csv = cleaned_data.to_csv(index=False).encode('utf-8')
+    st.download_button(
+        label="Download Cleaned Dataset",
+        data=cleaned_csv,
+        file_name="cleaned_dataset.csv",
+        mime="text/csv"
+    )
+
+    # Store the cleaned dataframe in session state for use in other pages
+    st.session_state['df'] = cleaned_data
 
-# Session State Access on Other Pages
-if 'df' in st.session_state:
-    data = st.session_state['df']
-    st.write("Dataset available for further analysis.")
 else:
-    st.warning("No dataset found. Please upload data to proceed.")
+    st.info("No dataset found in session. Please upload a CSV file on the previous page.")