Predicting stock prices using AI is a powerful application of deep learning. In this tutorial, we’ll walk through a complete project that uses LSTM (Long Short-Term Memory) networks to predict the future price of a stock, using TensorFlow and Keras.

---

✅ Step 1: Install Required Libraries

Install the required Python libraries using pip:

pip install yfinance numpy pandas scikit-learn matplotlib tensorflow

---

📁 Step 2: Import the Required Libraries

import yfinance as yf
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.preprocessing import MinMaxScaler
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, Dropout

---

📈 Step 3: Load Historical Stock Data

We’ll use Yahoo Finance to fetch historical closing prices of Apple (AAPL).

df = yf.download('AAPL', start='2015-01-01', end='2023-12-31')
df = df[['Close']]
df.head()

---

🧼 Step 4: Data Preprocessing

Normalize the data and prepare it for training.

scaler = MinMaxScaler(feature_range=(0, 1))
scaled_data = scaler.fit_transform(df)

train_data = scaled_data[0:int(len(scaled_data) * 0.8), :]

def create_dataset(data, time_step=60):
    x, y = [], []
    for i in range(time_step, len(data)):
        x.append(data[i-time_step:i, 0])
        y.append(data[i, 0])
    return np.array(x), np.array(y)

X_train, y_train = create_dataset(train_data)

X_train = X_train.reshape((X_train.shape[0], X_train.shape[1], 1))

---

🧠 Step 5: Build the LSTM Model

Here’s how we define the LSTM-based neural network model:

model = Sequential()
model.add(LSTM(50, return_sequences=True, input_shape=(X_train.shape[1], 1)))
model.add(Dropout(0.2))
model.add(LSTM(50))
model.add(Dropout(0.2))
model.add(Dense(1))

model.compile(optimizer='adam', loss='mean_squared_error')
model.summary()

---

🔁 Step 6: Train the Model

Train the model using the historical stock price data.

model.fit(X_train, y_train, epochs=10, batch_size=32)

---

🧪 Step 7: Make Predictions on Test Data

Now test the model using the remaining data and see how well it performs.

test_data = scaled_data[int(len(scaled_data) * 0.8) - 60:]
X_test, y_test = create_dataset(test_data)
X_test = X_test.reshape((X_test.shape[0], X_test.shape[1], 1))

predictions = model.predict(X_test)
predictions = scaler.inverse_transform(predictions)
real_prices = scaler.inverse_transform(y_test.reshape(-1, 1))

---

📊 Step 8: Plot Actual vs Predicted Prices

Visualize the predicted stock prices against the actual prices.

plt.figure(figsize=(12,6))
plt.plot(real_prices, color='blue', label='Actual Price')
plt.plot(predictions, color='red', label='Predicted Price')
plt.title('Stock Price Prediction')
plt.xlabel('Time')
plt.ylabel('Price')
plt.legend()
plt.show()

---

🔮 Bonus: Predict the Next Day’s Price

Want to know the stock price for the next day?

last_60_days = scaled_data[-60:]
last_60_days = last_60_days.reshape((1, 60, 1))
next_price = model.predict(last_60_days)
print("Predicted next day's price:", scaler.inverse_transform(next_price))

---

🧠 Conclusion

This basic LSTM model gives a solid introduction to time series forecasting using TensorFlow. You can improve accuracy by:

• Including more features (Open, High, Low, Volume)
• Using more layers or training for more epochs
• Tuning hyperparameters