import streamlit as st
import pandas as pd
import numpy as np
from sklearn.linear_model import LinearRegression
import matplotlib.pyplot as plt
import random
import string

# Remove the Streamlit default layout
st.set_page_config(layout="wide")

# Load the stock data from the CSV files
data1 = pd.read_csv('Google_test_data.csv')
data2 = pd.read_csv('tesla.csv')

# Combine the datasets into a dictionary
datasets = {'Google': data1, 'Tesla': data2}

# Generate a random string for unique key generation
def random_string(length):
    letters = string.ascii_lowercase
    return ''.join(random.choice(letters) for i in range(length))

# Get the user's dataset selection
key = f'dataset_selection_{random_string(5)}'
selected_dataset = st.selectbox('Select Stock', list(datasets.keys()), key=key)

# Retrieve the selected dataset
selected_data = datasets[selected_dataset]

# Prepare the data for prediction
X = np.arange(len(selected_data)).reshape(-1, 1)
y = selected_data['Close']

# Train the linear regression model
model = LinearRegression()
model.fit(X, y)

# Predict the stock prices
predictions = model.predict(X)

# Create subplots
fig, axes = plt.subplots(3, 1, figsize=(10, 15))

# Plot 1: Actual and Predicted Stock Prices
axes[0].plot(selected_data['Date'], y, label='Actual')
axes[0].plot(selected_data['Date'], predictions, label='Predicted')
axes[0].set_xlabel('Date')
axes[0].set_ylabel('Stock Price')
axes[0].set_title(f'{selected_dataset} Stock Price Prediction')
axes[0].legend()

# Plot 2: Volume of Trades
axes[1].plot(selected_data['Date'], selected_data['Volume'])
axes[1].set_xlabel('Date')
axes[1].set_ylabel('Volume')
axes[1].set_title(f'{selected_dataset} Volume of Trades')

# Plot 3: Daily Percentage Change in Stock Prices
daily_returns = selected_data['Close'].pct_change() * 100
axes[2].plot(selected_data['Date'], daily_returns)
axes[2].set_xlabel('Date')
axes[2].set_ylabel('Percentage Change')
axes[2].set_title(f'{selected_dataset} Daily Percentage Change in Stock Prices')

# Add a submit button
if st.button('Submit'):
    # Get the updated dataset selection
    key = f'dataset_selection_{random_string(5)}'
    selected_dataset = st.selectbox('Select Stock', list(datasets.keys()), key=key)

    # Retrieve the updated dataset
    selected_data = datasets[selected_dataset]

    # Prepare the data for prediction
    X = np.arange(len(selected_data)).reshape(-1, 1)
    y = selected_data['Close']

    # Train the linear regression model
    model.fit(X, y)

    # Predict the stock prices
    predictions = model.predict(X)

    # Clear the existing plots
    for ax in axes:
        ax.clear()

    # Update the plots with the new dataset
    axes[0].plot(selected_data['Date'], y, label='Actual')
    axes[0].plot(selected_data['Date'], predictions, label='Predicted')
    axes[0].set_xlabel('Date')
    axes[0].set_ylabel('Stock Price')
    axes[0].set_title(f'{selected_dataset} Stock Price Prediction')
    axes[0].legend()

    axes[1].plot(selected_data['Date'], selected_data['Volume'])
    axes[1].set_xlabel('Date')
    axes[1].set_ylabel('Volume')
    axes[1].set_title(f'{selected_dataset} Volume of Trades')

    daily_returns = selected_data['Close'].pct_change() * 100
    axes[2].plot(selected_data['Date'], daily_returns)
    axes[2].set_xlabel('Date')
    axes[2].set_ylabel('Percentage Change')
    axes[2].set_title(f'{selected_dataset} Daily Percentage Change in Stock Prices')

# Display the graphs in Streamlit
st.pyplot(fig)