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
from st_aggrid import AgGrid, GridUpdateMode
from st_aggrid.grid_options_builder import GridOptionsBuilder

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 = pd.DataFrame(columns=col_names, index=range(num_rows))
        
        gd = GridOptionsBuilder.from_dataframe(data)
        gd.configure_default_column(editable=True)
        gridoptions = gd.build()
        
        grid_table = AgGrid(data, gridOptions=gridoptions, 
                            update_mode=GridUpdateMode.VALUE_CHANGED, 
                            height=400)
        
        return grid_table['data']
    return None

def preprocess_data(data):
    st.subheader("Data Preprocessing")
    
    # Handle missing values
    if data.isnull().sum().sum() > 0:
        st.write("Handling missing values:")
        for column in data.columns:
            if data[column].isnull().sum() > 0:
                method = st.selectbox(f"Choose method for {column}:", 
                                      ["Drop", "Fill with mean", "Fill with median", "Fill with mode"])
                if method == "Drop":
                    data = data.dropna(subset=[column])
                elif method == "Fill with mean":
                    data[column].fillna(data[column].mean(), inplace=True)
                elif method == "Fill with median":
                    data[column].fillna(data[column].median(), inplace=True)
                elif method == "Fill with mode":
                    data[column].fillna(data[column].mode()[0], inplace=True)
    
    # Convert data types
    for column in data.columns:
        if data[column].dtype == 'object':
            try:
                data[column] = pd.to_numeric(data[column])
                st.write(f"Converted {column} to numeric.")
            except ValueError:
                st.write(f"Kept {column} as categorical.")
    
    return data

def perform_analysis(data):
    st.header("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)

    # Box plots for numerical columns
    st.write("Box Plots:")
    for column in numeric_data.columns:
        fig, ax = plt.subplots()
        sns.boxplot(data=data, y=column, ax=ax)
        st.pyplot(fig)

    # Bar plots for categorical columns
    categorical_columns = data.select_dtypes(include=['object']).columns
    if not categorical_columns.empty:
        st.write("Bar Plots for Categorical Variables:")
        for column in categorical_columns:
            fig, ax = plt.subplots()
            data[column].value_counts().plot(kind='bar', ax=ax)
            plt.title(f"Distribution of {column}")
            plt.xlabel(column)
            plt.ylabel("Count")
            st.pyplot(fig)

def main():
    st.title("Interactive EDA Toolkit")

    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())

        data = preprocess_data(data)
        perform_analysis(data)

if __name__ == "__main__":
    main()