src/streamlit_app.py

import streamlit as st
import pandas as pd
import numpy as np
import plotly.express as px
from scipy import stats
import io

# Metadata
AUTHOR = "Eduardo Nacimiento García"
EMAIL = "enacimie@ull.edu.es"
LICENSE = "Apache 2.0"

# Page config
st.set_page_config(
    page_title="SimpleStats",
    page_icon="📊",
    layout="wide",
    initial_sidebar_state="expanded",
)

# Title and credits
st.title("📊 SimpleStats")
st.markdown(f"**Author:** {AUTHOR} | **Email:** {EMAIL} | **License:** {LICENSE}")

st.write("""
Upload a CSV file or try the built-in demo dataset to perform statistical analysis: summary, charts, hypothesis tests, and more.
""")

# Generate demo dataset
@st.cache_data
def create_demo_data():
    np.random.seed(42)
    n = 200
    data = {
        "Age": np.random.normal(35, 12, n).astype(int),
        "Income": np.random.normal(45000, 15000, n),
        "Satisfaction": np.random.randint(1, 11, n),  # scale 1-10
        "Group": np.random.choice(["A", "B", "C"], n),
        "Gender": np.random.choice(["M", "F"], n, p=[0.6, 0.4]),
        "Purchase": np.random.choice([0, 1], n, p=[0.7, 0.3])
    }
    df = pd.DataFrame(data)
    # Introduce some nulls for demo
    df.loc[np.random.choice(df.index, 10), "Income"] = np.nan
    df.loc[np.random.choice(df.index, 5), "Age"] = np.nan
    return df

demo_df = create_demo_data()

# Button to load demo data
if st.button("🧪 Load Demo Dataset"):
    st.session_state['uploaded_file'] = demo_df.to_csv(index=False).encode('utf-8')
    st.session_state['file_name'] = "demo_data.csv"
    st.success("✅ Demo dataset loaded. Explore the features!")

# File uploader
uploaded_file = st.file_uploader("📂 Upload your CSV file", type=["csv"])

# Determine data source
if 'uploaded_file' in st.session_state and not uploaded_file:
    # Use demo if no file uploaded
    csv_bytes = st.session_state['uploaded_file']
    file_name = st.session_state['file_name']
    df = pd.read_csv(io.BytesIO(csv_bytes))
    st.info(f"Using demo dataset: `{file_name}`")
elif uploaded_file is not None:
    df = pd.read_csv(uploaded_file)
    st.success("✅ File uploaded successfully.")
else:
    df = None

if df is not None:
    # Show data preview
    with st.expander("🔍 Data Preview (first 10 rows)"):
        st.dataframe(df.head(10))

    # Basic info
    st.subheader("📌 Dataset Information")
    col1, col2, col3 = st.columns(3)
    col1.metric("Rows", df.shape[0])
    col2.metric("Columns", df.shape[1])
    col3.metric("Missing Values", df.isnull().sum().sum())

    # Identify numeric and categorical columns
    numeric_cols = df.select_dtypes(include=[np.number]).columns.tolist()
    categorical_cols = df.select_dtypes(include=['object', 'category']).columns.tolist()

    if len(numeric_cols) == 0:
        st.warning("⚠️ No numeric columns found for statistical analysis.")
    else:
        st.subheader("📈 Descriptive Statistics")
        st.dataframe(df[numeric_cols].describe())

        # Histogram
        st.subheader("📊 Histogram")
        selected_col = st.selectbox("Select a numeric column for histogram:", numeric_cols)
        fig_hist = px.histogram(df, x=selected_col, nbins=30, title=f"Histogram of {selected_col}", marginal="box")
        st.plotly_chart(fig_hist, use_container_width=True)

        # Scatter plot
        if len(numeric_cols) >= 2:
            st.subheader("📉 Scatter Plot")
            col_x = st.selectbox("Select X-axis column:", numeric_cols, key="x")
            col_y = st.selectbox("Select Y-axis column:", numeric_cols, key="y")
            color_by = st.selectbox("Color by (optional):", [None] + categorical_cols, key="color")
            if col_x != col_y:
                fig_scatter = px.scatter(df, x=col_x, y=col_y, color=color_by, title=f"{col_x} vs {col_y}")
                st.plotly_chart(fig_scatter, use_container_width=True)
            else:
                st.warning("⚠️ Please select two different columns.")

        # Correlation matrix
        st.subheader("🔗 Correlation Matrix")
        corr = df[numeric_cols].corr()
        fig_corr = px.imshow(corr, text_auto=".2f", aspect="auto", title="Correlation Matrix", color_continuous_scale='RdBu_r')
        st.plotly_chart(fig_corr, use_container_width=True)

        # Missing values per column
        st.subheader("🕳️ Missing Values by Column")
        nulls = df.isnull().sum()
        if nulls.sum() > 0:
            fig_nulls = px.bar(nulls, title="Missing Values by Column", labels={'value': 'Count', 'index': 'Column'}, color=nulls)
            st.plotly_chart(fig_nulls, use_container_width=True)
        else:
            st.success("✅ No missing values in the dataset.")

    # === STATISTICAL TESTS ===
    st.header("🧪 Statistical Tests")

    # Independent T-Test (for 2 groups)
    if len(numeric_cols) > 0 and len(categorical_cols) > 0:
        st.subheader("Independent T-Test (2 groups)")
        t_num_col = st.selectbox("Numeric variable:", numeric_cols, key="t_num")
        t_cat_col = st.selectbox("Categorical variable (must have exactly 2 groups):", 
                                [col for col in categorical_cols if df[col].nunique() == 2], 
                                key="t_cat")
        if t_cat_col:
            groups = df[t_cat_col].unique()
            if len(groups) == 2:
                group1 = df[df[t_cat_col] == groups[0]][t_num_col].dropna()
                group2 = df[df[t_cat_col] == groups[1]][t_num_col].dropna()
                t_stat, p_val = stats.ttest_ind(group1, group2, equal_var=False)
                st.write(f"**T-Test result between `{groups[0]}` and `{groups[1]}` for `{t_num_col}`:**")
                st.write(f"- **T-statistic:** {t_stat:.4f}")
                st.write(f"- **P-value:** {p_val:.4f}")
                if p_val < 0.05:
                    st.success("🟢 Statistically significant difference (p < 0.05)")
                else:
                    st.error("🔴 No statistically significant difference (p >= 0.05)")
            else:
                st.warning("Selected categorical variable does not have exactly 2 groups.")

    # ANOVA (for 3+ groups)
    if len(numeric_cols) > 0 and len(categorical_cols) > 0:
        st.subheader("ANOVA (3 or more groups)")
        anova_num_col = st.selectbox("Numeric variable:", numeric_cols, key="anova_num")
        anova_cat_col = st.selectbox("Categorical variable (3 or more groups):", 
                                    [col for col in categorical_cols if df[col].nunique() >= 3], 
                                    key="anova_cat")
        if anova_cat_col:
            groups = [df[df[anova_cat_col] == group][anova_num_col].dropna() for group in df[anova_cat_col].unique()]
            if len(groups) >= 3:
                f_stat, p_val = stats.f_oneway(*groups)
                st.write(f"**ANOVA result for `{anova_num_col}` grouped by `{anova_cat_col}`:**")
                st.write(f"- **F-statistic:** {f_stat:.4f}")
                st.write(f"- **P-value:** {p_val:.4f}")
                if p_val < 0.05:
                    st.success("🟢 At least one group is significantly different (p < 0.05)")
                else:
                    st.error("🔴 No significant differences between groups (p >= 0.05)")

    # Chi-Square Test (between two categorical variables)
    if len(categorical_cols) >= 2:
        st.subheader("Chi-Square Test (Association between categorical variables)")
        chi_col1 = st.selectbox("First categorical variable:", categorical_cols, key="chi1")
        chi_col2 = st.selectbox("Second categorical variable:", [col for col in categorical_cols if col != chi_col1], key="chi2")
        if chi_col1 and chi_col2:
            contingency_table = pd.crosstab(df[chi_col1], df[chi_col2])
            chi2, p_val, dof, expected = stats.chi2_contingency(contingency_table)
            st.write(f"**Chi-Square result between `{chi_col1}` and `{chi_col2}`:**")
            st.write(f"- **Chi² statistic:** {chi2:.4f}")
            st.write(f"- **P-value:** {p_val:.4f}")
            st.write(f"- **Degrees of freedom:** {dof}")
            if p_val < 0.05:
                st.success("🟢 Variables are associated (p < 0.05)")
            else:
                st.error("🔴 No evidence of association between variables (p >= 0.05)")
            with st.expander("📋 Contingency Table"):
                st.dataframe(contingency_table)

else:
    st.info("👆 Upload a CSV file or click 'Load Demo Dataset' to get started.")

# Footer
st.markdown("---")
st.caption(f"© {AUTHOR} | License {LICENSE} | Contact: {EMAIL}")