Upload 3 files

README_Synthsis.md

@@ -0,0 +1,32 @@
+readme_content = """---
+title: DataSynthis ML JobTask
+colorFrom: blue
+colorTo: green
+sdk: gradio
+sdk_version: 4.7.1
+app_file: app.py
+pinned: false
+---
+
+# Stock Price Forecasting
+
+DataSynthis ML Engineer Intern Job Task SET-B
+
+## Overview
+Time series forecasting for stock prices using ARIMA and LSTM models.
+
+## Models
+- ARIMA: Statistical approach
+- LSTM: Deep learning approach
+
+## Dataset
+Google (GOOGL) stock prices from 2018-2025 via Yahoo Finance
+
+## Performance
+ARIMA achieved RMSE of $3.48 vs LSTM's $30.88
+
+## Usage
+Enter stock ticker and select forecast horizon to generate predictions.
+
+with open('README.md', 'w') as f:
+    f.write(readme_content)

app_Synthsis.py

@@ -0,0 +1,64 @@
+import gradio as gr
+import numpy as np
+import pandas as pd
+from tensorflow.keras.models import load_model
+import joblib
+import yfinance as yf
+from datetime import datetime, timedelta
+import warnings
+warnings.filterwarnings('ignore')
+
+try:
+    lstm_model = load_model('lstm_googl_stock_model.h5')
+    scaler = joblib.load('scaler.pkl')
+except Exception as e:
+    print(f"Error: {e}")
+
+def predict_stock_price(ticker, days_ahead=30):
+    try:
+        end_date = datetime.now()
+        start_date = end_date - timedelta(days=365)
+        df = yf.download(ticker, start=start_date, end=end_date, progress=False)
+        
+        if df.empty or len(df) < 60:
+            return "Error: Insufficient data", None
+        
+        data = df[['Close']].values
+        scaled_data = scaler.transform(data)
+        lookback = 60
+        predictions = []
+        current_sequence = scaled_data[-lookback:].reshape(lookback, 1)
+        
+        for _ in range(days_ahead):
+            pred_scaled = lstm_model.predict(current_sequence.reshape(1, lookback, 1), verbose=0)
+            predictions.append(pred_scaled[0, 0])
+            current_sequence = np.append(current_sequence[1:], pred_scaled).reshape(lookback, 1)
+        
+        predictions_array = np.array(predictions).reshape(-1, 1)
+        predictions_actual = scaler.inverse_transform(predictions_array)
+        future_dates = pd.date_range(start=end_date + timedelta(days=1), periods=days_ahead)
+        results = pd.DataFrame({'Date': future_dates.strftime('%Y-%m-%d'), 'Predicted Price': predictions_actual.flatten().round(2)})
+        
+        last_price = data[-1][0]
+        avg_prediction = predictions_actual.mean()
+        change_pct = ((avg_prediction - last_price) / last_price * 100)
+        
+        summary = f"Stock: {ticker}\nCurrent: ${last_price:.2f}\nPredicted: ${avg_prediction:.2f}\nChange: {change_pct:+.2f}%"
+        return summary, results
+    except Exception as e:
+        return f"Error: {e}", None
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Stock Price Forecasting")
+    with gr.Row():
+        with gr.Column():
+            ticker_input = gr.Textbox(label="Stock Ticker", value="GOOGL")
+            days_slider = gr.Slider(1, 90, 30, label="Days")
+            predict_btn = gr.Button("Predict")
+        with gr.Column():
+            summary_output = gr.Textbox(label="Summary", lines=5)
+            table_output = gr.Dataframe(label="Predictions")
+    predict_btn.click(predict_stock_price, [ticker_input, days_slider], [summary_output, table_output])
+
+if __name__ == "__main__":
+    demo.launch()

requirements_Synthsis.txt

@@ -0,0 +1,7 @@
+tensorflow==2.15.0
+numpy==1.24.3
+pandas==2.0.3
+scikit-learn==1.3.0
+yfinance==0.2.38
+joblib==1.3.2
+gradio==4.7.1