Upload app.py

app.py

@@ -1,162 +1,108 @@
 import streamlit as st
-import plotly.express as px
 import pandas as pd
-import os
-import warnings
-
-warnings.filterwarnings('ignore')
-
-st.set_page_config(page_title="Superstore!!!", page_icon=":bar_chart:", layout="wide")
-
-st.title(" :bar_chart: Sample SuperStore EDA")
-st.markdown('<style>div.block-container{padding-top:1rem;}</style>', unsafe_allow_html=True)
-
-fl = st.file_uploader(":file_folder: Upload a file", type=(["csv", "txt", "xlsx", "xls"]))
-if fl is not None:
-    filename = fl.name
-    st.write(filename)
-    df = pd.read_csv(filename, encoding="ISO-8859-1", sep='\t')
-
-
-col1, col2 = st.columns((2))
-df["Order Date"] = pd.to_datetime(df["Order Date"])
-
-# Getting the min and max date
-startDate = pd.to_datetime(df["Order Date"]).min()
-endDate = pd.to_datetime(df["Order Date"]).max()
-
-with col1:
-    date1 = pd.to_datetime(st.date_input("Start Date", startDate))
-
-with col2:
-    date2 = pd.to_datetime(st.date_input("End Date", endDate))
-
-df = df[(df["Order Date"] >= date1) & (df["Order Date"] <= date2)].copy()
+import matplotlib.pyplot as plt
+import seaborn as sns
+import plotly.express as px
 
-st.sidebar.header("Choose your filter: ")
-# Create for Region
-region = st.sidebar.multiselect("Pick your Region", df["Region"].unique())
-if not region:
-    df2 = df.copy()
-else:
-    df2 = df[df["Region"].isin(region)]
+# Set page title
+st.set_page_config(page_title="Interactive BI Dashboard for Excel", layout="wide")
+st.title("Interactive Business Intelligence Dashboard for Excel")
+
+# Step 1: Upload Dataset
+st.sidebar.header('Upload your Excel file')
+uploaded_file = st.sidebar.file_uploader("Choose an Excel file", type=["xlsx", "xls"])
+
+# Step 2: Data Processing
+if uploaded_file is not None:
+    # Load the data from the Excel file
+    xls = pd.ExcelFile(uploaded_file)
+
+    # Show sheet names for easy sheet selection
+    sheet_names = xls.sheet_names
+    selected_sheet = st.sidebar.selectbox("Select a Sheet", sheet_names)
+
+    # Read data from the selected sheet
+    data = pd.read_excel(uploaded_file, sheet_name=selected_sheet)
+
+    # Display dataset overview
+    st.write(f"### Dataset Overview - {selected_sheet}")
+    st.write(data.head())  # Show first few rows
+
+    # Display dataset statistics
+    st.write("### Dataset Statistics")
+    st.write(data.describe())
+
+    # Display column names for easy column selection
+    columns = data.columns.tolist()
+
+    # Step 3: Dynamic Data Exploration and Visualization
+    st.sidebar.header('Data Exploration')
+
+    # Select feature for analysis
+    column_name = st.sidebar.selectbox("Select a column to analyze", columns)
+
+    # Filter numerical data for correlation heatmap
+    numeric_columns = data.select_dtypes(include='number').columns.tolist()
+    correlation_column = st.sidebar.selectbox("Select column for Correlation Heatmap", numeric_columns)
+
+    # Sidebar options for visualizations
+    chart_type = st.sidebar.selectbox("Choose a Chart Type", ['Bar Chart', 'Pie Chart', 'Line Chart', 'Histogram'])
+
+    # Step 4: Interactive Plotting and Visualizations
+    if chart_type == 'Bar Chart':
+        st.write(f"### Bar Chart of {column_name}")
+        fig, ax = plt.subplots(figsize=(10, 6))
+        data[column_name].value_counts().plot(kind='bar', ax=ax, color='skyblue')
+        ax.set_title(f'Bar Chart of {column_name}')
+        st.pyplot(fig)
+
+    elif chart_type == 'Pie Chart':
+        st.write(f"### Pie Chart of {column_name}")
+        fig, ax = plt.subplots(figsize=(8, 8))
+        data[column_name].value_counts().plot(kind='pie', autopct='%1.1f%%', ax=ax,
+                                              colors=sns.color_palette("Set3", len(data[column_name].unique())))
+        ax.set_title(f'Pie Chart of {column_name}')
+        st.pyplot(fig)
+
+    elif chart_type == 'Line Chart':
+        if 'Date' in data.columns:
+            st.write("### Line Chart (Time Series Data)")
+            data['Date'] = pd.to_datetime(data['Date'])
+            data.set_index('Date', inplace=True)
+            st.line_chart(data[column_name])
+        else:
+            st.write("### Line Chart (No Date Column)")
+            st.warning(
+                "Your dataset doesn't contain a Date column. Please upload a dataset with a Date column for time-series analysis.")
+
+    elif chart_type == 'Histogram':
+        st.write(f"### Histogram of {column_name}")
+        fig, ax = plt.subplots(figsize=(10, 6))
+        sns.histplot(data[column_name], kde=True, color='orange', ax=ax)
+        ax.set_title(f'Histogram of {column_name}')
+        st.pyplot(fig)
+
+    # Step 5: Correlation Heatmap for Numeric Features
+    if st.sidebar.checkbox('Show Correlation Heatmap'):
+        st.write("### Correlation Heatmap")
+        numeric_data = data.select_dtypes(include='number')
+        corr_matrix = numeric_data.corr()
+        fig, ax = plt.subplots(figsize=(12, 8))
+        sns.heatmap(corr_matrix, annot=True, fmt='.2f', cmap='coolwarm', ax=ax, linewidths=0.5)
+        ax.set_title("Correlation Matrix of Numeric Features")
+        st.pyplot(fig)
+
+    # Step 6: Interactive Plot with Plotly (for better interactivity)
+    st.sidebar.header("Advanced Interactive Visualization with Plotly")
+    if st.sidebar.checkbox('Show Interactive Scatter Plot'):
+        st.write("### Interactive Scatter Plot")
+        if len(numeric_columns) >= 2:
+            x_axis = st.sidebar.selectbox("Select X-axis for Scatter Plot", numeric_columns)
+            y_axis = st.sidebar.selectbox("Select Y-axis for Scatter Plot", numeric_columns)
+            scatter_fig = px.scatter(data, x=x_axis, y=y_axis, title=f"Scatter Plot: {x_axis} vs {y_axis}")
+            st.plotly_chart(scatter_fig)
+        else:
+            st.warning("Not enough numeric columns to create a scatter plot.")
 
-# Create for State
-state = st.sidebar.multiselect("Pick the State", df2["State"].unique())
-if not state:
-    df3 = df2.copy()
-else:
-    df3 = df2[df2["State"].isin(state)]
-
-# Create for City
-city = st.sidebar.multiselect("Pick the City", df3["City"].unique())
-
-# Filter the data based on Region, State and City
-
-if not region and not state and not city:
-    filtered_df = df
-elif not state and not city:
-    filtered_df = df[df["Region"].isin(region)]
-elif not region and not city:
-    filtered_df = df[df["State"].isin(state)]
-elif state and city:
-    filtered_df = df3[df["State"].isin(state) & df3["City"].isin(city)]
-elif region and city:
-    filtered_df = df3[df["Region"].isin(region) & df3["City"].isin(city)]
-elif region and state:
-    filtered_df = df3[df["Region"].isin(region) & df3["State"].isin(state)]
-elif city:
-    filtered_df = df3[df3["City"].isin(city)]
 else:
-    filtered_df = df3[df3["Region"].isin(region) & df3["State"].isin(state) & df3["City"].isin(city)]
-
-category_df = filtered_df.groupby(by=["Category"], as_index=False)["Sales"].sum()
-
-with col1:
-    st.subheader("Category wise Sales")
-    fig = px.bar(category_df, x="Category", y="Sales", text=['${:,.2f}'.format(x) for x in category_df["Sales"]],
-                 template="seaborn")
-    st.plotly_chart(fig, use_container_width=True, height=200)
-
-with col2:
-    st.subheader("Region wise Sales")
-    fig = px.pie(filtered_df, values="Sales", names="Region", hole=0.5)
-    fig.update_traces(text=filtered_df["Region"], textposition="outside")
-    st.plotly_chart(fig, use_container_width=True)
-
-cl1, cl2 = st.columns((2))
-with cl1:
-    with st.expander("Category_ViewData"):
-        st.write(category_df.style.background_gradient(cmap="Blues"))
-        csv = category_df.to_csv(index=False).encode('utf-8')
-        st.download_button("Download Data", data=csv, file_name="Category.csv", mime="text/csv",
-                           help='Click here to download the data as a CSV file')
-
-with cl2:
-    with st.expander("Region_ViewData"):
-        region = filtered_df.groupby(by="Region", as_index=False)["Sales"].sum()
-        st.write(region.style.background_gradient(cmap="Oranges"))
-        csv = region.to_csv(index=False).encode('utf-8')
-        st.download_button("Download Data", data=csv, file_name="Region.csv", mime="text/csv",
-                           help='Click here to download the data as a CSV file')
-
-filtered_df["month_year"] = filtered_df["Order Date"].dt.to_period("M")
-st.subheader('Time Series Analysis')
-
-linechart = pd.DataFrame(
-    filtered_df.groupby(filtered_df["month_year"].dt.strftime("%Y : %b"))["Sales"].sum()).reset_index()
-fig2 = px.line(linechart, x="month_year", y="Sales", labels={"Sales": "Amount"}, height=500, width=1000,
-               template="gridon")
-st.plotly_chart(fig2, use_container_width=True)
-
-with st.expander("View Data of TimeSeries:"):
-    st.write(linechart.T.style.background_gradient(cmap="Blues"))
-    csv = linechart.to_csv(index=False).encode("utf-8")
-    st.download_button('Download Data', data=csv, file_name="TimeSeries.csv", mime='text/csv')
-
-# Create a treem based on Region, category, sub-Category
-st.subheader("Hierarchical view of Sales using TreeMap")
-fig3 = px.treemap(filtered_df, path=["Region", "Category", "Sub-Category"], values="Sales", hover_data=["Sales"],
-                  color="Sub-Category")
-fig3.update_layout(width=800, height=650)
-st.plotly_chart(fig3, use_container_width=True)
-
-chart1, chart2 = st.columns((2))
-with chart1:
-    st.subheader('Segment wise Sales')
-    fig = px.pie(filtered_df, values="Sales", names="Segment", template="plotly_dark")
-    fig.update_traces(text=filtered_df["Segment"], textposition="inside")
-    st.plotly_chart(fig, use_container_width=True)
-
-with chart2:
-    st.subheader('Category wise Sales')
-    fig = px.pie(filtered_df, values="Sales", names="Category", template="gridon")
-    fig.update_traces(text=filtered_df["Category"], textposition="inside")
-    st.plotly_chart(fig, use_container_width=True)
-
-import plotly.figure_factory as ff
-
-st.subheader(":point_right: Month wise Sub-Category Sales Summary")
-with st.expander("Summary_Table"):
-    df_sample = df[0:5][["Region", "State", "City", "Category", "Sales", "Profit", "Quantity"]]
-    fig = ff.create_table(df_sample, colorscale="Cividis")
-    st.plotly_chart(fig, use_container_width=True)
-
-    st.markdown("Month wise sub-Category Table")
-    filtered_df["month"] = filtered_df["Order Date"].dt.month_name()
-    sub_category_Year = pd.pivot_table(data=filtered_df, values="Sales", index=["Sub-Category"], columns="month")
-    st.write(sub_category_Year.style.background_gradient(cmap="Blues"))
-
-# Create a scatter plot
-data1 = px.scatter(filtered_df, x="Sales", y="Profit", size="Quantity")
-data1['layout'].update(title="Relationship between Sales and Profits using Scatter Plot.",
-                       titlefont=dict(size=20), xaxis=dict(title="Sales", titlefont=dict(size=19)),
-                       yaxis=dict(title="Profit", titlefont=dict(size=19)))
-st.plotly_chart(data1, use_container_width=True)
-
-with st.expander("View Data"):
-    st.write(filtered_df.iloc[:500, 1:20:2].style.background_gradient(cmap="Oranges"))
-
-# Download orginal DataSet
-csv = df.to_csv(index=False).encode('utf-8')
-st.download_button('Download Data', data=csv, file_name="Data.csv", mime="text/csv")
+    st.write("Please upload an Excel file to get started.")