correlation_plot

import streamlit as st
import pandas as pd
import altair as alt

url = "contract_awards_in_investment_project_financing_22-11-2024.csv"
df = pd.read_csv(url)

# Filter numeric columns for correlation analysis
numeric_columns = df.select_dtypes(include=['float64', 'int64']).columns
st.header("Correlation Analysis")

# Dropdowns for user selection of columns to correlate
col1 = st.selectbox("Select First Numeric Attribute", numeric_columns, index=0)
col2 = st.selectbox("Select Second Numeric Attribute", numeric_columns, index=1)

# Correlation calculation
correlation_value = df[col1].corr(df[col2])
st.write(f"**Correlation between {col1} and {col2}: {correlation_value:.2f}**")

# Create a scatter plot
correlation_plot = (
    alt.Chart(df)
    .mark_circle(size=60)
    .encode(
        x=alt.X(f"{col1}:Q", title=col1),
        y=alt.Y(f"{col2}:Q", title=col2),
        color="Region:N",  # Color by region for additional context
        tooltip=["Project Name", "Supplier Name", col1, col2],
    )
    .properties(
        title=f"Correlation Plot: {col1} vs. {col2}",
        width=600,
        height=400
    )
)
st.altair_chart(correlation_plot, use_container_width=True)