app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
from io import StringIO
import openpyxl

def load_data(file):
    file_extension = file.name.split('.')[-1].lower()
    if file_extension == 'csv':
        data = pd.read_csv(file)
    elif file_extension in ['xls', 'xlsx']:
        data = pd.read_excel(file)
    else:
        st.error("Unsupported file format. Please upload a CSV, XLS, or XLSX file.")
        return None
    return data

def manual_data_entry():
    st.subheader("Manual Data Entry")
    col_names = st.text_input("Enter column names separated by commas:").split(',')
    col_names = [name.strip() for name in col_names if name.strip()]
    
    if col_names:
        num_rows = st.number_input("Enter number of rows:", min_value=1, value=5)
        data = []
        for i in range(num_rows):
            row = []
            for col in col_names:
                value = st.text_input(f"Enter value for {col} (Row {i+1}):")
                row.append(value)
            data.append(row)
        
        return pd.DataFrame(data, columns=col_names)
    return None

def perform_analysis(data):
    st.header("4. Analysis")
    
    # EDA
    st.subheader("Exploratory Data Analysis")
    
    # Summary statistics
    st.write("Summary Statistics:")
    st.write(data.describe())

    # Correlation heatmap
    st.write("Correlation Heatmap:")
    numeric_data = data.select_dtypes(include=['float64', 'int64'])
    if not numeric_data.empty:
        fig, ax = plt.subplots(figsize=(10, 8))
        sns.heatmap(numeric_data.corr(), annot=True, cmap='coolwarm', ax=ax)
        st.pyplot(fig)
    else:
        st.write("No numeric columns available for correlation heatmap.")

    # Pairplot
    st.write("Pairplot:")
    if not numeric_data.empty:
        fig = sns.pairplot(numeric_data)
        st.pyplot(fig)
    else:
        st.write("No numeric columns available for pairplot.")

    # Histogram
    st.write("Histograms:")
    for column in numeric_data.columns:
        fig, ax = plt.subplots()
        sns.histplot(data[column], kde=True, ax=ax)
        st.pyplot(fig)

def main():
    st.title("PPDAC Data Analysis Toolkit")

    # Problem
    st.header("1. Problem")
    problem = st.text_area("Define your problem:")

    # Plan
    st.header("2. Plan")
    plan = st.text_area("Describe your plan:")

    # Data
    st.header("3. Data")
    data_input_method = st.radio("Choose data input method:", ("Upload File", "Manual Entry"))
    
    if data_input_method == "Upload File":
        uploaded_file = st.file_uploader("Choose a CSV, XLS, or XLSX file", type=["csv", "xls", "xlsx"])
        if uploaded_file is not None:
            data = load_data(uploaded_file)
        else:
            data = None
    else:
        data = manual_data_entry()
    
    if data is not None:
        st.write("Data Preview:")
        st.write(data.head())

        # Convert columns to numeric where possible
        for col in data.columns:
            try:
                data[col] = pd.to_numeric(data[col])
            except ValueError:
                pass  # Keep as non-numeric if conversion fails

        perform_analysis(data)

        # Conclusion
        st.header("5. Conclusion")
        conclusion = st.text_area("Write your conclusion based on the analysis:")

if __name__ == "__main__":
    main()