model.py

import yfinance as yf
import pandas as pd
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
import matplotlib.pyplot as plt
import io
import base64

# Function to fetch data and train model
def predict_price(symbol):
    df = yf.download(symbol, period="3mo", interval="1d")  # last 3 months
    if df.empty or len(df) < 60:
        return "❌ Not enough data to train model", None

    df = df[['Close']].dropna()
    df.reset_index(drop=True, inplace=True)

    # Create features: previous 5 days' close prices to predict next day
    X, y = [], []
    for i in range(5, len(df)-1):
        X.append(df['Close'].iloc[i-5:i].values)
        y.append(df['Close'].iloc[i+1])

    X = np.array(X)
    y = np.array(y)

    model = LinearRegression()
    model.fit(X, y)

    # Predict the next day's price
    last_5_days = df['Close'].iloc[-5:].values.reshape(1, -1)
    predicted = model.predict(last_5_days)[0]

    # Plot: Actual vs Predicted on last 10 days
    test_X = X[-10:]
    test_y = y[-10:]
    pred_y = model.predict(test_X)

    fig, ax = plt.subplots()
    ax.plot(range(10), test_y, label="Actual", marker='o')
    ax.plot(range(10), pred_y, label="Predicted", marker='x')
    ax.set_title(f"📈 Prediction vs Actual for {symbol}")
    ax.set_xlabel("Day")
    ax.set_ylabel("Price")
    ax.legend()

    return predicted, fig