Update app.py

app.py

@@ -1,18 +1,56 @@
-from datasets import load_dataset
-from tabulate import tabulate
 import gradio as gr
+import pandas as pd
+import plotly.express as px
 
-# Load CSV as Hugging Face Dataset
-def view_data():
-    dataset = load_dataset("csv", data_files="sample_data.csv")["train"]
-    table = tabulate(dataset[:5], headers="keys", tablefmt="github")
-    return table
+# Global variable to hold the dataset
+data = pd.DataFrame()
 
-with gr.Blocks() as demo:
-    gr.Markdown("### 📊 Sample Dataset Viewer")
-    view_btn = gr.Button("View First 5 Rows")
-    output = gr.Textbox(label="Dataset Preview", lines=10)
+def load_and_show(csv_file):
+    global data
+    if csv_file is None:
+        return "Please upload a CSV file.", None, None
+    data = pd.read_csv(csv_file.name)
+    summary = data.describe().reset_index()
     
-    view_btn.click(view_data, outputs=output)
+    return (
+        data.head().to_dict(orient="records"),  # Show first few rows in table
+        summary.to_dict(orient="records"),      # Show summary stats
+        None                                   # Empty plot initially
+    )
+
+def plot_column(col_name):
+    global data
+    if data.empty or col_name not in data.columns:
+        return None
+    if pd.api.types.is_numeric_dtype(data[col_name]):
+        fig = px.histogram(data, x=col_name, nbins=30, title=f"Distribution of {col_name}")
+        return fig
+    else:
+        return None
 
+with gr.Blocks() as demo:
+    gr.Markdown("# 📊 CSV Viewer with Search, Filter & Stats")
+    
+    with gr.Row():
+        csv_input = gr.File(label="Upload CSV", file_types=[".csv"])
+    
+    table = gr.DataFrame(headers=None, interactive=True, label="Dataset Preview", max_rows=10)
+    summary_table = gr.DataFrame(headers=None, interactive=False, label="Summary Statistics")
+    plot = gr.Plot()
+    
+    csv_input.change(load_and_show, inputs=csv_input, outputs=[table, summary_table, plot])
+    
+    col_dropdown = gr.Dropdown(choices=[], label="Select column to plot histogram")
+    col_dropdown.change(plot_column, inputs=col_dropdown, outputs=plot)
+    
+    # Update column dropdown choices after CSV load
+    def update_columns(csv_file):
+        if csv_file is None:
+            return []
+        df = pd.read_csv(csv_file.name)
+        numeric_cols = [c for c in df.columns if pd.api.types.is_numeric_dtype(df[c])]
+        return numeric_cols
+    
+    csv_input.change(update_columns, inputs=csv_input, outputs=col_dropdown)
+    
 demo.launch()