import gradio as gr
import numpy as np
import pandas as pd
import yfinance as yf
from datetime import datetime
from tensorflow.keras.models import load_model
from joblib import load
import io
from PIL import Image

# Load the saved LSTM model and scaler
lstm_model = load_model('lstm_model.h5')
scaler = load('scaler.joblib')

# Define the list of stocks
stock_list = ['GOOG', 'AAPL', 'TSLA', 'AMZN', 'MSFT']

# Function to get the last row of stock data
def get_last_stock_data(ticker):
    try:
        start_date = '2010-01-01'
        end_date = datetime.now().strftime('%Y-%m-%d')
        data = yf.download(ticker, start=start_date, end=end_date)
        last_row = data.iloc[-1]
        return last_row.to_dict()
    except Exception as e:
        return str(e)

# Function to make predictions
def predict_stock_price(ticker, open_price, high_price, low_price, close_price, adj_close_price, volume):
    try:
        start_date = '2010-01-01'
        end_date = datetime.now().strftime('%Y-%m-%d')
        data = yf.download(ticker, start=start_date, end=end_date)

        # Prepare the data
        data = data[['Close']]
        dataset = data.values
        scaled_data = scaler.transform(dataset)

        # Append the user inputs as the last row in the data
        user_input = np.array([[close_price]])
        user_input_scaled = scaler.transform(user_input)
        scaled_data = np.vstack([scaled_data, user_input_scaled])

        # Prepare the data for LSTM
        x_test_lstm = []
        for i in range(60, len(scaled_data)):
            x_test_lstm.append(scaled_data[i-60:i])

        x_test_lstm = np.array(x_test_lstm)
        x_test_lstm = np.reshape(x_test_lstm, (x_test_lstm.shape[0], x_test_lstm.shape[1], 1))

        # LSTM Predictions
        lstm_predictions = lstm_model.predict(x_test_lstm)
        lstm_predictions = scaler.inverse_transform(lstm_predictions)
        next_day_lstm_price = lstm_predictions[-1][0]
        
        # Plot the data
        plt.figure(figsize=(10, 6))
        plt.plot(data.index, data['Close'], label='Historical Close Prices')
        plt.axvline(x=data.index[-1], color='r', linestyle='--', label='Prediction Date')
        plt.plot([data.index[-1], data.index[-1] + pd.DateOffset(1)], [data['Close'].iloc[-1], next_day_lstm_price], 'go-', label='Predicted Price')
        plt.title(f'Predicted Closing Price for {ticker}')
        plt.xlabel('Date')
        plt.ylabel('Close Price USD ($)')
        plt.legend()

        # Save the plot to a buffer
        buf = io.BytesIO()
        plt.savefig(buf, format='png')
        buf.seek(0)
        plt.close()
        
      # Load the image from the buffer
        img = Image.open(buf)
        
        result = f"Predicted future price for {ticker}: ${next_day_lstm_price:.2f}"

        return result, img
    except Exception as e:
        return str(e)

# Set up Gradio interface
ticker_input = gr.Dropdown(choices=stock_list, label="Stock Ticker")

def get_user_inputs(ticker):
    last_data = get_last_stock_data(ticker)
    if isinstance(last_data, str):
        return gr.Textbox.update(value=last_data)
    else:
        return gr.update(inputs=[
            gr.Number(value=last_data['Open'], label='Open'),
            gr.Number(value=last_data['High'], label='High'),
            gr.Number(value=last_data['Low'], label='Low'),
            gr.Number(value=last_data['Close'], label='Close'),
            gr.Number(value=last_data['Adj Close'], label='Adj Close'),
            gr.Number(value=last_data['Volume'], label='Volume')
        ])

iface = gr.Interface(
    fn=predict_stock_price,
    inputs=[
        ticker_input,
        gr.Number(label="Open"),
        gr.Number(label="High"),
        gr.Number(label="Low"),
        gr.Number(label="Close"),
        gr.Number(label="Adj Close"),
        gr.Number(label="Volume")
    ],
    outputs=[gr.Textbox(), gr.Image(type="pil")],
    title="Stock Price Predictor",
    description="Select the stock ticker and input the last recorded values to predict the closing price using the LSTM model."
)

iface.launch()