app.py

import gradio as gr
import pandas as pd
import sweetviz as sv
import tempfile
import os
import category_encoders as ce
import umap
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler
from autoviz.AutoViz_Class import AutoViz_Class
import shutil
import warnings
import io
import base64
warnings.filterwarnings('ignore')

class DataAnalyzer:
    def __init__(self):
        self.temp_dir = tempfile.mkdtemp()
        self.df = None
        self.AV = AutoViz_Class()
        
    def generate_sweetviz_report(self, df):
        if df is None:
            return "Please upload a dataset first"
        
        self.df = df
        report = sv.analyze(df)
        report_path = os.path.join(self.temp_dir, "report.html")
        report.show_html(report_path, open_browser=False)
        
        with open(report_path, 'r', encoding='utf-8') as f:
            html_content = f.read()
        
        html_with_table = f"""
        <table width="100%" style="border-collapse: collapse;">
            <tr>
                <td style="padding: 20px; border: 1px solid #ddd;">
                    <div style="height: 800px; overflow: auto;">
                        {html_content}
                    </div>
                </td>
            </tr>
        </table>
        """
        
        os.remove(report_path)
        return html_with_table

    def preprocess_dataframe(self, df):
        df = df.copy()
        
        # Convert 'value' column to numeric if possible
        if 'value' in df.columns:
            df['value'] = pd.to_numeric(df['value'].replace('[\$,]', '', regex=True), errors='coerce')
        
        # Handle datetime columns
        for col in df.columns:
            if df[col].dtype == 'object':
                try:
                    df[col] = pd.to_datetime(df[col], errors='ignore')
                except:
                    pass
        
        # Convert categorical columns with low cardinality
        for col in df.select_dtypes(include=['object']).columns:
            if df[col].nunique() < 50:
                df[col] = df[col].astype('category')
        
        return df

    def generate_autoviz_report(self, df):
        if df is None:
            return "Please upload a dataset first"
            
        viz_temp_dir = os.path.join(self.temp_dir, "autoviz_output")
        if os.path.exists(viz_temp_dir):
            shutil.rmtree(viz_temp_dir)
        os.makedirs(viz_temp_dir)

        try:
            # Preprocess the dataframe
            df = self.preprocess_dataframe(df)
            
            # Sample if needed
            if len(df) > 5000:
                df = df.sample(n=5000, random_state=42)
            
            print("\nDataset Info:")
            print(df.info())
            print("\nColumn Types:")
            print(df.dtypes)
            
            plt.close('all')
            
            # Create a directory for plots
            plots_dir = os.path.join(viz_temp_dir, "plots")
            os.makedirs(plots_dir, exist_ok=True)
            
            # Run AutoViz
            dfte = self.AV.AutoViz(
                filename='',
                sep=',',
                depVar='value',  # Set value as target variable
                dfte=df,
                header=0,
                verbose=1,
                lowess=False,
                chart_format='html',
                max_rows_analyzed=5000,
                max_cols_analyzed=30,
                save_plot_dir=plots_dir
            )

            # Generate summary statistics
            numeric_cols = df.select_dtypes(include=['number']).columns
            categorical_cols = df.select_dtypes(include=['category', 'object']).columns
            
            numeric_stats = df[numeric_cols].describe().round(2) if len(numeric_cols) > 0 else pd.DataFrame()
            categorical_stats = df[categorical_cols].describe() if len(categorical_cols) > 0 else pd.DataFrame()
            
            # Create HTML content with styling
            html_content = """
            <style>
                .table {
                    width: 100%;
                    margin-bottom: 1rem;
                    color: #212529;
                    border-collapse: collapse;
                }
                .table-striped tbody tr:nth-of-type(odd) {
                    background-color: rgba(0,0,0,.05);
                }
                .table td, .table th {
                    padding: .75rem;
                    border: 1px solid #dee2e6;
                }
                .table th {
                    background-color: #f8f9fa;
                }
                pre {
                    background-color: #f8f9fa;
                    padding: 1rem;
                    border-radius: 4px;
                }
                .viz-container {
                    margin: 20px 0;
                    padding: 20px;
                    border: 1px solid #ddd;
                    border-radius: 5px;
                }
            </style>
            """
            
            html_content += f"""
            <div class="viz-container">
                <h2 style="text-align: center;">Data Analysis Report</h2>
                
                <div style="margin: 20px;">
                    <h3>Dataset Overview</h3>
                    <p>Total Rows: {len(df)}</p>
                    <p>Total Columns: {len(df.columns)}</p>
                    
                    <h3>Numeric Variables Summary</h3>
                    <div style="overflow-x: auto;">
                        {numeric_stats.to_html(classes='table table-striped')}
                    </div>
                    
                    <h3>Categorical Variables Summary</h3>
                    <div style="overflow-x: auto;">
                        {categorical_stats.to_html(classes='table table-striped')}
                    </div>
                    
                    <h3>Column Types</h3>
                    <pre>{df.dtypes.to_string()}</pre>
                </div>
            """

            # Add plots if they exist
            if os.path.exists(plots_dir):
                for file in sorted(os.listdir(plots_dir)):
                    if file.endswith('.html'):
                        with open(os.path.join(plots_dir, file), 'r', encoding='utf-8') as f:
                            plot_content = f.read()
                            if plot_content.strip():
                                html_content += f"""
                                <div class="viz-container">
                                    <h3>{file.replace('.html', '').replace('_', ' ').title()}</h3>
                                    {plot_content}
                                </div>
                                """

            html_content += "</div>"
            return html_content

        except Exception as e:
            import traceback
            error_message = f"""
            <div style="padding: 20px; border: 1px solid red; border-radius: 5px;">
                <h3>Error in AutoViz Analysis</h3>
                <p>Error details: {str(e)}</p>
                <p>Stack trace:</p>
                <pre>{traceback.format_exc()}</pre>
                <p>Dataset Info:</p>
                <pre>
                Rows: {len(df)}
                Columns: {len(df.columns)}
                Types:\n{df.dtypes.to_string()}
                </pre>
            </div>
            """
            return error_message
        finally:
            if os.path.exists(viz_temp_dir):
                shutil.rmtree(viz_temp_dir)


def create_interface():
    analyzer = DataAnalyzer()
    
    with gr.Blocks(theme=gr.themes.Soft()) as demo:
        gr.Markdown("""
        # Data Analysis Dashboard
        This dashboard provides comprehensive data analysis and visualization capabilities.
        """)
        
        # Store the dataframe in a state variable
        current_df = gr.State(None)
        
        with gr.Tabs():
            # First Tab: Data Upload & Preview
            with gr.TabItem("Data Upload & Preview"):
                with gr.Row():
                    with gr.Column(scale=2):
                        file_input = gr.File(
                            label="Upload CSV File",
                            file_types=[".csv"],
                            file_count="single"
                        )
                    with gr.Column(scale=1):
                        gr.Markdown("""
                        ### Upload Instructions
                        1. Select a CSV file
                        2. File will be automatically loaded
                        3. Preview will appear below
                        """)
                
                with gr.Row():
                    data_info = gr.Markdown("No data uploaded yet")
                
                with gr.Row():
                    data_preview = gr.Dataframe(
                        label="Data Preview",
                        interactive=False,
                        wrap=True
                    )
                
                def load_data(file):
                    if file is None:
                        return "No data uploaded yet", None, None
                    try:
                        df = pd.read_csv(file.name)
                        info_text = f"""
                        ### Dataset Information
                        - Rows: {len(df)}
                        - Columns: {len(df.columns)}
                        - Memory Usage: {df.memory_usage(deep=True).sum() / 1024:.2f} KB
                        - Column Types: {dict(df.dtypes.value_counts())}
                        """
                        return info_text, df.head(10), df
                    except Exception as e:
                        return f"Error loading file: {str(e)}", None, None
                
                file_input.change(
                    fn=load_data,
                    inputs=[file_input],
                    outputs=[data_info, data_preview, current_df]
                )
            
            # Second Tab: Sweetviz Analysis
            with gr.TabItem("Sweetviz Analysis"):
                with gr.Row():
                    with gr.Column(scale=2):
                        sweetviz_button = gr.Button(
                            "Generate Sweetviz Report",
                            variant="primary"
                        )
                    with gr.Column(scale=1):
                        gr.Markdown("""
                        ### Sweetviz Analysis Features
                        - Comprehensive data profiling
                        - Statistical analysis
                        - Feature correlations
                        - Missing value analysis
                        """)
                
                with gr.Row():
                    sweetviz_output = gr.HTML(
                        label="Sweetviz Report",
                        value="Click the button above to generate the report"
                    )
                
                def generate_sweetviz(df):
                    if df is None:
                        return "Please upload a dataset first"
                    try:
                        return analyzer.generate_sweetviz_report(df)
                    except Exception as e:
                        return f"Error generating Sweetviz report: {str(e)}"
                
                sweetviz_button.click(
                    fn=generate_sweetviz,
                    inputs=[current_df],
                    outputs=[sweetviz_output]
                )
            
            # Third Tab: AutoViz Analysis
            with gr.TabItem("AutoViz Analysis"):
                with gr.Row():
                    with gr.Column(scale=2):
                        autoviz_button = gr.Button(
                            "Generate AutoViz Report",
                            variant="primary"
                        )
                    with gr.Column(scale=1):
                        gr.Markdown("""
                        ### AutoViz Analysis Features
                        - Automated visualization generation
                        - Distribution analysis
                        - Correlation plots
                        - Feature relationships
                        - Time series analysis (if applicable)
                        """)
                
                with gr.Row():
                    autoviz_output = gr.HTML(
                        label="AutoViz Report",
                        value="Click the button above to generate the report"
                    )
                
                def generate_autoviz(df):
                    if df is None:
                        return "Please upload a dataset first"
                    try:
                        return analyzer.generate_autoviz_report(df)
                    except Exception as e:
                        return f"Error generating AutoViz report: {str(e)}"
                
                autoviz_button.click(
                    fn=generate_autoviz,
                    inputs=[current_df],
                    outputs=[autoviz_output]
                )

    return demo

if __name__ == "__main__":
    demo = create_interface()
    demo.launch(
        server_name="0.0.0.0",
        server_port=7860,
        show_error=True,
        share=False  # Set to True if you want to create a public link
    )