project_2.py

import pandas as pd
import streamlit as st
import plotly.express as px

# Classy and centered title
st.markdown(
    """
    <h1 style="text-align: center; color: skyblue;"> Tips Dataset Dashboard</h1>
    """,
    unsafe_allow_html=True
)

# File Upload
file = st.file_uploader("📁 Upload a CSV file", type=["csv"])

if file is not None:
    df = pd.read_csv(file)
    
    st.markdown("### 🔍 Preview of Data")
    st.write(df.head())

    # Separate Numerical and Categorical Columns
    num = df.select_dtypes('number')
    cat = df.select_dtypes('object')
    numerical = num.columns.tolist()
    categorical = cat.columns.tolist()

    st.markdown("### 🟨 Categorical Features")
    st.write(cat)

    st.markdown("### 🟦 Numerical Features")
    st.write(num)

    # 📊 Categorical Feature Analysis
    st.markdown("## 📋 Categorical Feature Analysis")
    for col in categorical:
        st.write(f"### 🔸 Feature: `{col}`")
        st.write("🔢 Value Counts:")
        st.write(df[col].value_counts())
        st.write("📜 Unique Values:")
        st.write(df[col].unique())
        st.write("🧮 Number of Unique Values:", df[col].nunique())
        st.write("❓ Missing Values:", df[col].isnull().sum())
        st.markdown("---")

    # 📈 Numerical Feature Analysis
    st.markdown("## 📊 Numerical Feature Analysis")
    for col in numerical:
        st.write(f"### 🔹 Feature: `{col}`")
        st.write("📏 Mean:", df[col].mean())
        st.write("🪙 Median:", df[col].median())
        st.write("📉 Std Deviation:", df[col].std())
        st.write("📐 Variance:", df[col].var())
        st.write("📈 Skewness:", df[col].skew())
        st.write("🔺 Kurtosis:", df[col].kurt())
        st.write("🔽 Minimum:", df[col].min())
        st.write("🔼 Maximum:", df[col].max())
        st.write("❓ Missing Values:", df[col].isnull().sum())
        st.markdown("---")

    # 🔗 Bivariate Analysis
    st.markdown("## 🔗 Bivariate Analysis")

    # Scatter Plot
    st.markdown("### 📍 Scatter Plot")
    num_x = st.selectbox('🧭 Select X-axis (Numerical)', numerical, key='scatter_x')
    num_y = st.selectbox('📌 Select Y-axis (Numerical)', numerical, key='scatter_y')
    if num_x and num_y:
        st.write(f"🔎 Scatter plot between `{num_x}` and `{num_y}`")
        fig = px.scatter(df, x=num_x, y=num_y, title=f'Scatter plot: {num_x} vs {num_y}')
        st.plotly_chart(fig)

    # Box Plot
    st.markdown("### 🎁 Box Plot")
    cat_feature = st.selectbox('🧊 Select Category', categorical, key='box_cat')
    num_feature = st.selectbox('📐 Select Value (Numerical)', numerical, key='box_num')
    if cat_feature and num_feature:
        fig = px.box(df, x=cat_feature, y=num_feature, title=f'Box plot of {num_feature} by {cat_feature}')
        st.plotly_chart(fig)

    # Correlation Matrix
    st.markdown("### 🔥 Correlation Matrix")
    corr = num.corr()
    fig = px.imshow(corr, text_auto=True, title='Correlation Heatmap')
    st.plotly_chart(fig)

    # Pairplot
    st.markdown("### 🌐 Pairplot")
    if len(numerical) > 1:
        fig = px.scatter_matrix(df, dimensions=numerical, title='Pairplot of Numerical Features')
        st.plotly_chart(fig)

    # Count Plot
    st.markdown("### 📊 Count Plot")
    cat_feature_count = st.selectbox('📋 Select a categorical feature', categorical, key='count_plot')
    if cat_feature_count:
        value_counts_df = df[cat_feature_count].value_counts().reset_index()
        value_counts_df.columns = [cat_feature_count, 'Count']
        fig = px.bar(value_counts_df, x=cat_feature_count, y='Count',
                     title=f'Count Plot for {cat_feature_count}',
                     labels={cat_feature_count: cat_feature_count, 'Count': 'Count'})
        st.plotly_chart(fig)

    # Distribution Plot
    st.markdown("### 🧮 Distribution Plot")
    num_feature_dist = st.selectbox('🔢 Select a numerical feature', numerical, key='dist_plot')
    if num_feature_dist:
        fig = px.histogram(df, x=num_feature_dist, nbins=30,
                           title=f'Distribution of {num_feature_dist}')
        st.plotly_chart(fig)

    # Pie Chart
    st.markdown("### 🥧 Pie Chart")
    cat_feature_pie = st.selectbox('🧠 Select categorical feature for pie chart', categorical, key='pie_chart')
    if cat_feature_pie:
        fig = px.pie(df, names=cat_feature_pie, title=f'Pie Chart of {cat_feature_pie}')
        st.plotly_chart(fig)

else:
    st.warning("📁 Please upload a CSV file to begin.")