Update app.py

app.py

@@ -1,85 +1,101 @@
 import gradio as gr
 import pandas as pd
+import numpy as np
 import seaborn as sns
 import matplotlib.pyplot as plt
 import io
+import os
+from sklearn.decomposition import PCA
+from sklearn.preprocessing import StandardScaler
 from docx import Document
-from docx.shared import Inches
 
-def generate_report(df):
-    # Create Word Document
-    doc = Document()
-    doc.add_heading('Amrita Manthan: Preprocessing Report', 0)
-    doc.add_paragraph(f"Author: Prof. B. Santhi, SRC, SASTRA")
+def perform_eda_and_preprocess(file):
+    if file is None: return [None]*6
     
-    # Data Description
-    doc.add_heading('1. Data Description', level=1)
-    description = df.describe().to_string()
-    doc.add_paragraph(description)
+    df = pd.read_csv(file.name)
+    numeric_df = df.select_dtypes(include=[np.number])
     
-    # Missing Values
-    doc.add_heading('2. Preprocessing: Missing Values', level=1)
-    null_data = df.isnull().sum().to_string()
-    doc.add_paragraph(null_data)
+    # 1. Preprocessing: Null/Missing Checks
+    null_info = df.isnull().sum().to_frame(name="Missing Values").reset_index()
+    df_clean = df.fillna(df.mean(numeric_only=True)) # Simple Imputation
     
-    report_path = "preprocessing_report.docx"
-    doc.save(report_path)
-    return report_path
-
-def process_data(file):
-    if file is None:
-        return None, "Please upload a CSV file.", None, None
+    # 2. Statistical Features
+    stats = df.describe(include='all').reset_index()
     
-    # Load Data
-    df = pd.read_csv(file.name)
+    # 3. PCA (Principal Component Analysis)
+    pca_result = "Not enough numeric data"
+    if numeric_df.shape[1] >= 2:
+        x = StandardScaler().fit_transform(numeric_df.dropna())
+        pca = PCA(n_components=2)
+        components = pca.fit_transform(x)
+        pca_df = pd.DataFrame(data=components, columns=['PC1', 'PC2'])
+        pca_result = f"Explained Variance: {pca.explained_variance_ratio_}"
+    
+    # 4. Generate 15+ Plots
+    fig, axes = plt.subplots(5, 3, figsize=(20, 25))
+    axes = axes.flatten()
+    cols = numeric_df.columns[:15] # Take first 15 numeric columns
     
-    # 1. EDA: Basic Stats
-    desc = df.describe(include='all').reset_index()
+    for i, col in enumerate(cols):
+        sns.histplot(df[col], kde=True, ax=axes[i], color='skyblue')
+        axes[i].set_title(f'Dist: {col}')
     
-    # 2. EDA: Correlation Plot
-    plt.figure(figsize=(10, 6))
-    sns.heatmap(df.corr(numeric_only=True), annot=True, cmap='coolwarm')
     plt.tight_layout()
-    plot_path = "eda_plot.png"
+    plot_path = "multi_plot_grid.png"
     plt.savefig(plot_path)
     plt.close()
-    
-    # 3. Generate Docx
-    report = generate_report(df)
-    
-    return df.head(10), desc, plot_path, report
 
-# --- Gradio UI Design ---
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    # Header Section
-    with gr.Column(elem_id="header"):
-        gr.Image("sastra_logo.jpg", width=150, container=False) # Ensure file exists
-        gr.Markdown("<h1 style='text-align: center;'>Amrita Manthan</h1>")
-        gr.Markdown("<h3 style='text-align: center;'>Preprocessing & Data Mining Portal</h3>")
-        gr.Markdown("<p style='text-align: center;'><i>Prof. B. Santhi, SRC, SASTRA</i></p>")
+    # 5. Generate Professional DOCX Report
+    doc_path = "Data_Mining_Report.docx"
+    doc = Document()
+    doc.add_heading('Amrita Manthan - Preprocessing Report', 0)
+    doc.add_paragraph("Institution: SRC, SASTRA | Faculty: Prof. B. Santhi")
+    doc.add_heading('Missing Value Analysis', level=1)
+    doc.add_paragraph(null_info.to_string())
+    doc.add_heading('PCA Summary', level=1)
+    doc.add_paragraph(str(pca_result))
+    doc.save(doc_path)
+
+    return df.head(10), stats, null_info, plot_path, pca_result, doc_path
 
-    with gr.Tabs():
-        with gr.TabItem("Step 1: Upload & Process"):
-            file_input = gr.File(label="Upload CSV Dataset", file_types=[".csv"])
-            process_btn = gr.Button("Analyze Data", variant="primary")
+# --- UI Layout ---
+with gr.Blocks(theme=gr.themes.Default()) as demo:
+    # Header with Logo and Blue Bold Centered Title
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Image("logo.jpg", width=100, show_label=False, container=False)
+        with gr.Column(scale=4):
+            gr.Markdown(
+                "<h1 style='text-align: center; color: #003366;'><b>Amrita Manthan</b></h1>"
+                "<h2 style='text-align: center; color: #003366;'><b>Preprocessing & Data Mining Portal</b></h2>"
+                "<p style='text-align: center;'><b>Prof. B. Santhi, SRC, SASTRA</b></p>"
+            )
+
+    with gr.Tabs() as ribbon:
+        with gr.TabItem("📁 Data Upload"):
+            file_input = gr.File(label="Upload CSV")
+            process_btn = gr.Button("START MINING", variant="primary")
             
-        with gr.TabItem("Step 2: Data View"):
+        with gr.TabItem("📊 Statistics & Nulls"):
             with gr.Row():
-                out_raw = gr.DataFrame(label="Raw Data (First 10 Rows)")
-                out_desc = gr.DataFrame(label="Statistical Summary")
-                
-        with gr.TabItem("Step 3: Visual EDA"):
-            out_plot = gr.Image(label="Correlation Heatmap")
+                out_raw = gr.DataFrame(label="Dataset Preview")
+                out_null = gr.DataFrame(label="Missing Value Check")
+            out_stat = gr.DataFrame(label="Statistical Summary")
+
+        with gr.TabItem("📉 15-Plot EDA Grid"):
+            out_plot = gr.Image(label="Distribution Analysis (15 Plots)")
             
-        with gr.TabItem("Step 4: Download Report"):
-            gr.Markdown("Click below to download the full preprocessing documentation.")
-            out_file = gr.File(label="Download DOCX Report")
+        with gr.TabItem("🧬 PCA & Feature Selection"):
+            pca_out = gr.Textbox(label="PCA Explained Variance")
+            gr.Markdown("Note: PCA reduces dimensionality to 2 Principal Components for visualization.")
+
+        with gr.TabItem("📥 Download Report"):
+            out_file = gr.File(label="Download DOCX Documentation")
 
-    # Click Logic
     process_btn.click(
-        fn=process_data,
+        fn=perform_eda_and_preprocess,
         inputs=[file_input],
-        outputs=[out_raw, out_desc, out_plot, out_file]
+        outputs=[out_raw, out_stat, out_null, out_plot, pca_out, out_file]
     )
 
 demo.launch()