app.py

import gradio as gr
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import os

# ------------------ Data Cleaning ------------------
def clean_data(file_path, remove_na, fill_na, method, remove_duplicates, outlier, convert_dtype, dtype_column, dtype):
    if file_path is None:
        return None, None

    # Read file
    if file_path.endswith(".csv"):
        df = pd.read_csv(file_path)
    else:
        df = pd.read_excel(file_path)

    # Cleaning operations
    if remove_na:
        df = df.dropna()

    if fill_na:
        if method == "Mean":
            df = df.fillna(df.mean(numeric_only=True))
        elif method == "Median":
            df = df.fillna(df.median(numeric_only=True))
        else:
            df = df.fillna(df.mode().iloc[0])

    if remove_duplicates:
        df = df.drop_duplicates()

    if outlier:
        Q1 = df.quantile(0.25)
        Q3 = df.quantile(0.75)
        IQR = Q3 - Q1
        df = df[~((df < (Q1 - 1.5 * IQR)) | (df > (Q3 + 1.5 * IQR))).any(axis=1)]

    if convert_dtype and dtype_column:
        try:
            if dtype == "int":
                df[dtype_column] = df[dtype_column].astype(int)
            elif dtype == "float":
                df[dtype_column] = df[dtype_column].astype(float)
            else:
                df[dtype_column] = df[dtype_column].astype(str)
        except Exception as e:
            return None, None

    # Save cleaned file
    output_path = "cleaned_output.csv"
    df.to_csv(output_path, index=False)

    # Generate dashboard images
    dashboard_images = generate_dashboard(df)

    return output_path, dashboard_images

# ------------------ Dashboard Analytics ------------------
def generate_dashboard(df):
    os.makedirs("dashboard_images", exist_ok=True)
    images = []

    # Missing value heatmap
    plt.figure(figsize=(6,4))
    sns.heatmap(df.isna(), cbar=False)
    plt.title("Missing Values Heatmap")
    heatmap_file = "dashboard_images/missing_heatmap.png"
    plt.savefig(heatmap_file)
    plt.close()
    images.append(heatmap_file)

    # Numeric histograms
    numeric_cols = df.select_dtypes(include=np.number).columns
    for col in numeric_cols:
        plt.figure(figsize=(5,3))
        sns.histplot(df[col].dropna(), kde=True)
        plt.title(f"Histogram of {col}")
        hist_file = f"dashboard_images/hist_{col}.png"
        plt.savefig(hist_file)
        plt.close()
        images.append(hist_file)

    return images

# ------------------ Guide Popup ------------------
def show_guide_popup():
    guide_text = """
## 🧹 Data Cleaning Tool Guide

### 1️⃣ Basic Statistical Terms
- **Mean (Average)**: Sum of all numeric values divided by the number of values.
- **Median**: Middle value when sorted.
- **Mode**: Most frequent value.

### 2️⃣ Cleaning Options
- **Remove Missing Values**: Deletes rows with missing values.
- **Fill Missing Values**: Replace NaNs with Mean/Median/Mode.
- **Remove Duplicates**: Deletes repeated rows.
- **Outlier Removal (IQR Method)**: Removes extreme values beyond 1.5*IQR.
- **Convert Column Data Type**: Change a column to int, float, or string.

### 3️⃣ Dashboard Analytics
- **Missing Values Heatmap**: See which columns have missing data.
- **Numeric Histograms**: Check distribution of numeric columns.

### 4️⃣ Recommended Workflow
1. Upload CSV/Excel.
2. Remove/fill missing values.
3. Remove duplicates/outliers.
4. Convert column types.
5. Explore dashboard.
6. Download cleaned dataset.

**Tip:** Always keep a backup of the original file!
"""
    return guide_text

# ------------------ Update Columns for Dtype ------------------
def update_columns(file_path):
    if file_path is None:
        return gr.update(choices=[], value=None)
    if file_path.endswith(".csv"):
        df = pd.read_csv(file_path)
    else:
        df = pd.read_excel(file_path)
    cols = list(df.columns)
    return gr.update(choices=cols, value=cols[0] if cols else None)

# ------------------ Gradio Interface ------------------
with gr.Blocks() as demo:
    gr.Markdown("## 🧹 Data Cleaning Tool with Dashboard Analytics")

    with gr.Row():
        file_input = gr.File(label="Upload CSV or Excel", type="filepath")
        guide_button = gr.Button("Show Guide")

    remove_na = gr.Checkbox(label="Remove rows with missing values")
    fill_na = gr.Checkbox(label="Fill missing values")
    method = gr.Dropdown(["Mean", "Median", "Mode"], label="Fill method", value="Mean")
    remove_duplicates = gr.Checkbox(label="Remove duplicate rows")
    outlier = gr.Checkbox(label="Remove outliers (IQR Method)")
    convert_dtype = gr.Checkbox(label="Convert column data type")

    dtype_column = gr.Dropdown([], label="Column to convert")
    dtype = gr.Dropdown(["int", "float", "string"], label="Convert to", value="int")

    file_input.change(update_columns, inputs=file_input, outputs=dtype_column)

    clean_button = gr.Button("Apply Cleaning")
    cleaned_output = gr.File(label="Download Cleaned Data")
    dashboard_output = gr.Gallery(label="Dashboard Analytics", show_label=True)

    # Guide popup (hidden initially)
    guide_popup = gr.Textbox(value="", label="Guide", interactive=False, visible=False, lines=25)

    # Show guide on button click
    def show_guide_and_popup():
        return gr.update(value=show_guide_popup(), visible=True)

    guide_button.click(
        show_guide_and_popup,
        inputs=None,
        outputs=guide_popup
    )

    # Cleaning function
    clean_button.click(
        clean_data,
        inputs=[file_input, remove_na, fill_na, method, remove_duplicates, outlier, convert_dtype, dtype_column, dtype],
        outputs=[cleaned_output, dashboard_output]
    )

demo.launch()