Popular Libraries

Import the tensorflow library.

from AlgorithmImports import *
import tensorflow as tf

In the Initializeinitialize method, subscribe to some data so you can train the tensorflow model and make predictions.

# Subscribe to security data and store symbol for referencing in the algorithm.
self._symbol = self.add_equity("SPY", Resolution.DAILY).symbol

In this example, build a neural-network regression prediction model that uses the following features and labels:

The following image shows the time difference between the features and labels:

Follow these steps to create a method to build the model:

1. Define some parameters, including the number of factors, neurons, and epochs.

# Set hyperparameters: number of input features, neurons in each hidden layer to capture complexity, training epochs for sufficient learning, and learning rate for optimization control.
num_factors = 5
num_neurons_1 = 10
num_neurons_2 = 10
num_neurons_3 = 5
self.epochs = 20
self.learning_rate = 0.0001

2. Create the model using built-in Keras API.

# Define the model with sequential layers: input layer with ReLU activation to introduce non-linearity, two hidden layers for feature extraction, and an output layer for predictions.
self.model = tf.keras.sequential([
    tf.keras.layers.dense(num_neurons_1, activation=tf.nn.relu, input_shape=(num_factors,)),  # input shape required
    tf.keras.layers.dense(num_neurons_2, activation=tf.nn.relu),
    tf.keras.layers.dense(num_neurons_3, activation=tf.nn.relu),
    tf.keras.layers.dense(1)
])

You can train the model at the beginning of your algorithm and you can periodically re-train it as the algorithm executes.

Warm Up Training Data

You need historical data to initially train the model at the start of your algorithm. To get the initial training data, in the Initializeinitialize method, make a history request.

# Initialize training data with a rolling window of size 300 days.
training_length = 300
self.training_data = RollingWindow[float](training_length)
history = self.history[TradeBar](self._symbol, training_length, Resolution.DAILY)
for trade_bar in history:
    self.training_data.add(trade_bar.close)

Define a Training Method

To train the model, define a method that fits the model with the training data.

# Prepare feature and label data for training by processing rolling window data to create time-series sequences for model training.
def get_features_and_labels(self, lookback=5):
    lookback_series = []

    data = pd.Series(list(self.training_data)[::-1])
    for i in range(1, lookback + 1):
        df = data.diff(i)[lookback:-1]
        df.name = f"close-{i}"
        lookback_series.append(df)

    X = pd.concat(lookback_series, axis=1).reset_index(drop=True).dropna()
    Y = data.diff(-1)[lookback:-1].reset_index(drop=True)
    return X.values, Y.values

def my_training_method(self):
    features, labels = self.get_features_and_labels()

    # Define the loss function, we use MSE in this example
    def loss_mse(target_y, predicted_y):
        return tf.reduce_mean(tf.square(target_y - predicted_y))

    # Train the model
    optimizer = tf.keras.optimizers.adam(learning_rate=self.learning_rate)
    for i in range(self.epochs):
        with tf.gradient_tape() as t:
            loss = loss_mse(labels, self.model(features))

        jac = t.gradient(loss, self.model.trainable_weights)
        optimizer.apply_gradients(zip(jac, self.model.trainable_weights))

Set Training Schedule

To train the model at the beginning of your algorithm, in the Initializeinitialize method, call the Traintrain method.

# Train the model initially to provide a baseline for prediction and decision-making.
self.train(self.my_training_method)

To periodically re-train the model as your algorithm executes, in the Initializeinitialize method, schedule some training sessions.

# Train the model every Sunday at 8:00 AM
self.train(self.date_rules.every(DayOfWeek.SUNDAY), self.time_rules.at(8, 0), self.my_training_method)

Update Training Data

To update the training data as the algorithm executes, in the OnDataon_data method, add the current close price to the RollingWindow that holds the training data.

# Add the latest bar to training data to ensure the model is trained with the most recent market data.
def on_data(self, slice: Slice) -> None:
    if self._symbol in slice.bars:
        self.training_data.add(slice.bars[self._symbol].close)

To predict the labels of new data, in the OnDataon_data method, get the most recent set of features and then call the run method with new features.

# Generate feature set and predict with the latest data for current market decisions.
new_features, __ = self.get_features_and_labels()
prediction = self.model(new_features)
prediction = float(prediction.numpy()[-1])

You can use the label prediction to place orders.

# Use label prediction to place orders based on forecasted market direction.
if prediction > 0:
    self.SetHoldings(self._symbol, 1)
elif prediction < 0:
    self.SetHoldings(self._symbol, -1)

Follow these steps to save Tensorflow models into the Object Store:

1. Set the key name of the model to be stored in the Object Store.

# Set the key to store the model in Object Store for reuse across sessions.
model_key = "model.keras"

Note that the model has to have the suffix .keras.

2. Call the GetFilePathget_file_path method with the key.

# Get the file path to correctly save and access the model in Object Store.
file_name = self.object_store.get_file_path(model_key)

This method returns the file path where the model will be stored.

3. Call the save method with the model and file path.

# Serialize Python objects into a file to save the model's state for other runs.
model.save(file_name)

4. Save the model to the file path.

# Save the model to object store to capture model state for reusability.
self.object_store.save(model_key)

You can load and trade with pre-trained tensorflow models that you saved in the Object Store. To load a tensorflow model from the Object Store, in the Initializeinitialize method, get the file path to the saved model and then recall the graph and weights of the model.

# Load the tensorflow model from Object Store to use its saved state and update with new data if needed.
def initialize(self) -> None:
    model_key = 'model.keras'
    if self.object_store.contains_key(model_key):
        file_name = self.object_store.get_file_path(model_key)
        self.model = tf.keras.models.load_model(file_name)

The ContainsKeycontains_key method returns a boolean that represents if the model.keras is in the Object Store. If the Object Store doesn't contain the keys, save the model using them before you proceed.