
Explore building deep neural networks for forex trading, first in Excel and then in mql5, covering forward pass, backpropagation, gradient descent, and min-max data normalization with live signal visualization.
Learn to perform backpropagation in the output layer by computing how weights and a bias affect the cost, using sigmoid derivatives to update weights and reduce loss.
Build a deep neural network in mql5 by creating an expert advisor, configuring inputs and chart appearance, and implementing initialization and on-tick logic with new-bar detection.
Declare and manage c tensor objects for weights, bias, inputs, and outputs, check validity, and delete them to prepare tensors for neural network data flow.
Get data from macd histogram and rsi indicators, set ema 12/26 and signal 9, configure rsi 18, and build a 5000-path, 3-column training matrix for bullish or bearish candlestick prediction.
Develop a live trading pipeline that runs on each new price quote, building an inputs vector from macd and rsi indicators using copy buffer.
Split the data matrix into training and testing sets by implementing a randomize function with seed 42, using Fisher's algorithm to shuffle rows and ensure reproducible training.
Create a custom copy function to copy from one vector to another in MQL5, using copy from, copy to, start from, and total, returning a boolean and resizing the destination.
Implement a data split function that randomizes the dataset and partitions it into x_train, y_train, x_test, y_test with a 0.7 training ratio to train a neural network and prevent overfitting.
Configure hidden layers by declaring the number of layers and neurons per layer, storing sizes in a vector (e.g., 15, 10, 7). Warn against overfitting in forex time series.
Split 70% training and 30% testing data, normalize inputs with min max normalization, train the neural network with back propagation, then evaluate with a confusion matrix.
The forex market, a dynamic beast fueled by a complex web of data, demands a keen eye for intricate patterns and the agility to adapt. While traditional methods have served us well, their limitations become apparent in this ever-evolving landscape. These limitations have been overcome by the development of Deep Neural Networks (DNNs), a revolutionary approach poised to transform the forex trading landscape.
In the previous courses on Neural networks, we only managed to create perceptrons, by taking input data, passing it through activation functions and getting outputs or predictions. These basically did hidden layer calculations without actual hidden layers. Hence we can liken perceptrons to single or lone traders who analyze charts, limited by its ability to do more work and specialize. Deep NNs, however, are like a collaborative team of analysts. With Information flowing from one expert analyst to another through their layered architecture, each layer building upon the insights or work done by the previous one. It's like a team identifying specific puzzle pieces, working together to reveal the bigger picture with remarkable clarity.
This layered structure empowers DNNs to tackle problems that would leave a single perceptron incompetent. They excel at unveiling hidden trends, sifting through vast datasets and uncovering hidden correlations and patterns that escape the human eye. A DNN, by analyzing a broader range of data points and their relationships, could potentially identify small data relationships and adjust predictions accordingly.
This course builds on your existing knowledge of neural networks to take you on a deep dive into Deep Neural Networks (DNNs) for forex trading. You'll learn to code your own DNNs using MQL5, a programming language for MetaTrader platforms.
We'll start with a basic DNN built in Excel, providing a foundation for the more complex coding in MQL5. This hands-on exercise will focus on a real-world forex trading problem, showcasing the practical applications of DNNs. In Excel, you'll explore the forward pass, backpropagation for updating weights and biases, and gradient descent for training the network.
Equipped with this knowledge, you'll transition to coding DNNs in MQL5. We'll cover designing the network, setting general parameters, and constructing a tensor for data storage. You'll learn to collect and prepare input data, including randomization and normalization. The course will then guide you through the forward pass and backpropagation in MQL5, along with extracting signals and visualizing predictions. By the end, you'll not only understand the theory of DNNs but also be able to code them for forex trading.
So what are you waiting for? Click hard on that enroll button now and join us in this wonderful journey of coding a deep neural network in MQL5