app.py

import pandas as pd
import ydata_profiling
import gradio as gr
from pydantic_settings import BaseSettings
from tempfile import NamedTemporaryFile
import sweetviz as sv
#from datatile.summary.df import DataFrameSummary
from autoviz.AutoViz_Class import AutoViz_Class
from traceml.summary.df import DataFrameSummary


def generate_report(file, type):
    df = pd.read_csv(file) if file.name.endswith(".csv") else pd.read_excel(file)

    pandas_html_report =ydata_profiling.ProfileReport(df).to_html() 
    temp_file1 = NamedTemporaryFile(delete=False, suffix=".html")
    temp_file1.write(pandas_html_report.encode('utf-8'))
    temp_file1.close()
    print('Pandas completed')
    
    # dataprep_report = create_report(df)
    # temp_file2 = NamedTemporaryFile(delete=False, suffix=".html")
    # temp_file2.write(dataprep_report.encode('utf-8'))
    # temp_file2.close()

    def convert_mixed_to_str(df):
  
      mixed_cols = df.select_dtypes(include=['object']).columns  # Detect object dtype (usually mixed)
      for col in mixed_cols:
        df[col] = df[col].astype(str)  # Convert mixed columns to strings
      return df

    df = convert_mixed_to_str(df.copy())

    sweetviz_report = sv.analyze(df)
    #sweetviz_report=sweetviz_report.show_html()
    #print(type(sweetviz_report))
    report=sweetviz_report.show_html(  filepath='SWEETVIZ_REPORT.html', 
            open_browser=False, 
            layout='widescreen', 
            scale=None)
    HTMLFileToBeOpened = open('SWEETVIZ_REPORT.html', "r") 
  
    # Reading the file and storing in a variable 
    contents = HTMLFileToBeOpened.read() 
    print('Pandas completed')
    
  
    temp_file3 = NamedTemporaryFile(delete=False, suffix=".html")
    temp_file3.write(contents.encode('utf-8'))
    temp_file3.close()

    #AUTOVIZ
    AV = AutoViz_Class()

    dfviz = AV.AutoViz(
    "",
    sep=",",
    depVar="",
    dfte=df,
    header=0,
    verbose=2,
    lowess=False,
    chart_format="html",
    max_rows_analyzed=500,
    max_cols_analyzed=20,
    save_plot_dir="plots"
    )

    
    
   
    dfs = DataFrameSummary(df)
    sd=dfs.summary()
    sd.index.name = 'Parameters'
    sd1= sd.reset_index(drop=False)
    
    return temp_file1.name  ,temp_file3.name ,dfviz,sd1
    
     
with gr.Blocks() as cluster:
    with gr.Column():
        
        with gr.Row():
            file=gr.File(file_types=['.csv', '.xlsx'], label="Upload a CSV or Excel file")
            btn=gr.Button(value="Download Report") 
            
        with gr.Row():
            
            
            gr.HTML(value="""<h1 style="color: #3399FF; text-shadow: 1px 1px 2px #ddd;">PANDAS REPORT</h1>""")
            out1=gr.File(label="Download CSV")
            gr.HTML(value="""<h1 style="color: #3399FF; text-shadow: 1px 1px 2px #ddd;">DATAPREP REPORT</h1>""")
            out2=gr.File(label="Download CSV")
            gr.HTML(value="""<h1 style="color: #3399FF; text-shadow: 1px 1px 2px #ddd;">SWEETVIZ REPORT</h1>""")
            out3=gr.File(label="Download CSV")
        with gr.Row():
            with gr.Column():
                gr.Marker("Uploaded File")
                dataframe1=gr.Dataframe()
            with gr.Column():
                gr.Marker("Columns Analysis")
                dataframe2=gr.Dataframe()
    btn.click(generate_report,inputs=[file],outputs=[out1,out3,dataframe1,dataframe2])
cluster.launch()