import yfinance as yf
import pandas as pd
import numpy as np
import plotly.graph_objs as go
from prophet import Prophet
import gradio as gr

# Function to fetch stock data
def fetch_stock_data(ticker, start_date, end_date):
    stock_data = yf.download(ticker, start=start_date, end=end_date)
    return stock_data

# Train the model using Prophet
def train_prophet_model(stock_data):
    df = stock_data.reset_index()[['Date', 'Close']]
    df.columns = ['ds', 'y']
    
    model = Prophet()
    model.fit(df)
    
    future = model.make_future_dataframe(periods=90)  # Predict for the next 3 months
    forecast = model.predict(future)
    return forecast

# Function to display stock information and prediction
def predict_stock(ticker, start_date, end_date):
    # Fetch stock data
    stock_data = fetch_stock_data(ticker, start_date, end_date)
    
    # Train the model and get predictions
    forecast = train_prophet_model(stock_data)
    
    # Calculate metrics: % change, highest, lowest
    start_price = stock_data['Close'].iloc[0]
    current_price = stock_data['Close'].iloc[-1]
    percentage_change = ((current_price - start_price) / start_price) * 100
    highest_price = stock_data['Close'].max()
    lowest_price = stock_data['Close'].min()
    
    # Buy/Sell prediction (if price is predicted to increase, suggest Buy)
    predicted_next_value = forecast['yhat'].iloc[-1]
    suggestion = "Buy" if predicted_next_value > current_price else "Sell"
    
    # Plot historical and predicted prices
    fig = go.Figure()
    fig.add_trace(go.Scatter(x=stock_data.index, y=stock_data['Close'], mode='lines', name='Historical'))
    fig.add_trace(go.Scatter(x=forecast['ds'], y=forecast['yhat'], mode='lines', name='Predicted'))
    
    return fig, percentage_change, highest_price, lowest_price, suggestion

# Define Gradio UI components
tickers = ['AAPL', 'GOOG', 'MSFT', 'AMZN', 'TSLA', 'META', 'NVDA', 'NFLX', 'BABA', 'ORCL']
start_date = gr.components.DatePicker(label="Start Date", value="2023-01-01")
end_date = gr.components.DatePicker(label="End Date", value="2023-09-30")
ticker_dropdown = gr.components.Dropdown(choices=tickers, label="Stock Ticker")

# Gradio function to wrap the prediction
def gradio_interface(ticker, start_date, end_date):
    fig, percentage_change, highest_price, lowest_price, suggestion = predict_stock(ticker, start_date, end_date)
    return fig, f"{percentage_change:.2f}%", f"{highest_price:.2f}", f"{lowest_price:.2f}", suggestion

# Gradio interface and launch
output = [gr.components.Plot(label="Stock Price Plot"), gr.components.Textbox(label="Percentage Change"),
          gr.components.Textbox(label="Highest Price"), gr.components.Textbox(label="Lowest Price"),
          gr.components.Textbox(label="Prediction (Buy/Sell)")]

demo = gr.Interface(fn=gradio_interface, inputs=[ticker_dropdown, start_date, end_date], outputs=output, title="Stock Predictor App")

demo.launch()