Update app.py

app.py

@@ -3,75 +3,91 @@ import pandas as pd
 import numpy as np
 from scipy.fft import fft, fftfreq
 from sklearn.preprocessing import MinMaxScaler
-from tensorflow.keras.models import Sequential, load_model
-import requests
+from tensorflow.keras.models import Sequential
+from tensorflow.keras.layers import LSTM, Dense
+import matplotlib.pyplot as plt
 
-# --- Pre-trained Model (Simple LSTM) ---
-def build_model():
-    model = Sequential([
-        tf.keras.layers.LSTM(32, input_shape=(30, 1)),
-        tf.keras.layers.Dense(1)
-    ])
-    model.compile(loss='mse', optimizer='adam')
-    return model
+def load_data(input_source):
+    """Handle both uploaded files and URLs"""
+    if isinstance(input_source, str) and input_source.startswith("http"):
+        # Load from URL
+        df = pd.read_csv(
+            input_source,
+            engine='python',
+            on_bad_lines='warn',
+            encoding='utf-8'
+        )
+    else:
+        # Load from uploaded file
+        df = pd.read_csv(
+            input_source.name,
+            engine='python',
+            on_bad_lines='warn',
+            encoding='utf-8'
+        )
+    
+    # Common cleaning steps
+    df = df.drop(columns=['Province/State', 'Lat', 'Long'], errors='ignore')
+    df = df.groupby('Country/Region').sum().T
+    df.index = pd.to_datetime(df.index)
+    df['Global'] = df.sum(axis=1)
+    return df['Global'].diff().fillna(0)
 
-# --- Core Functions ---
-def analyze_data(data_url, prediction_days=30):
+def analyze_data(input_source):
     try:
-        # 1. Fetch data
-        df = pd.read_csv(data_url)
-        dates = df.columns[4:]  # COVID data format
-        values = df.drop(columns=['Province/State', 'Country/Region', 'Lat', 'Long']).sum(axis=0)[4:].values.astype(float)
+        data = load_data(input_source)
         
-        # 2. Detect cycles
-        N = len(values)
-        yf = fft(values)
+        # Analysis logic
+        N = len(data)
+        yf = fft(data.values)
         xf = fftfreq(N, 1)[:N//2]
-        dominant_freq = xf[np.argmax(np.abs(yf[0:N//2]))]
-        cycle_days = int(1/dominant_freq)
-        
-        # 3. Make predictions (simplified)
-        scaler = MinMaxScaler()
-        scaled = scaler.fit_transform(values.reshape(-1, 1))
-        
-        model = build_model()
-        model.fit(scaled[:-10], scaled[10:], epochs=5, verbose=0)  # Quick training
-        
-        preds = model.predict(scaled[-30:].reshape(1, 30, 1))
-        preds = scaler.inverse_transform(preds).flatten().tolist()
+        cycle_days = int(1/xf[np.argmax(np.abs(yf[0:N//2]))])
         
-        # 4. Generate insights
-        insights = [
-            f"Dominant cycle: {cycle_days} days",
-            f"Next {prediction_days}-day trend: {'↑ Upward' if preds[-1] > preds[0] else '↓ Downward'}",
-            "Action: Monitor closely around cycle peaks"
-        ]
+        # Create plot
+        fig, ax = plt.subplots()
+        ax.plot(data.index, data.values)
+        ax.set_title("COVID-19 Daily New Cases Analysis")
         
-        # Simple plot
-        plot = pd.DataFrame({
-            'Historical': values,
-            'Predicted': [None]*(len(values)) + preds
-        }).plot(title="Cases Analysis").figure
+        return (
+            f"🔮 Analysis Results:\n"
+            f"- Cycle: {cycle_days} days\n"
+            f"- Latest 30-day average: {data[-30:].mean():.1f} cases/day\n"
+            f"- Current trend: {'↑ Rising' if data[-1] > data[-7] else '↓ Falling'}",
+            fig
+        )
         
-        return plot, insights
-    
     except Exception as e:
-        return None, [f"Error: {str(e)}"]
+        return f"❌ Error: {str(e)}", None
 
-# --- Gradio Interface ---
-interface = gr.Interface(
-    fn=analyze_data,
-    inputs=[
-        gr.Textbox(label="Data URL", 
-                   value="https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/data/time_series_covid19_confirmed_global.csv"),
-        gr.Number(label="Days to Predict", value=30)
-    ],
-    outputs=[
-        gr.Plot(label="Analysis"),
-        gr.JSON(label="Insights")
-    ],
-    title="DeepSeek Lite Analyzer",
-    description="Analyze time-series data from public URLs. Works best with COVID-19 format data."
-)
+# Create hybrid interface with chat and file upload
+with gr.Blocks(theme=gr.themes.Soft()) as app:
+    gr.Markdown("# 📊 Data Analysis Bot")
+    gr.Markdown("Upload a CSV file or paste a COVID data URL")
+    
+    with gr.Row():
+        with gr.Column():
+            file_upload = gr.File(label="Upload CSV", file_count=1)
+            url_input = gr.Textbox(label="Or paste URL here")
+            submit_btn = gr.Button("Analyze")
+        with gr.Column():
+            chat = gr.Chatbot(height=400)
+            plot_output = gr.Plot()
+    
+    # Handle both input methods
+    submit_btn.click(
+        fn=analyze_data,
+        inputs=[gr.combine(file_upload, url_input)],
+        outputs=[chat, plot_output]
+    )
+    
+    # Example inputs
+    gr.Examples(
+        examples=[
+            ["https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/data/time_series_covid19_confirmed_global.csv"],
+            ["sample_data.csv"]  # Upload this via Hugging Face
+        ],
+        inputs=[url_input]
+    )
 
-interface.launch()
+if __name__ == "__main__":
+    app.launch()