Intel Neuron Bot EA MT5

Intel Neuron Bot EA MT5 is a configurable neural-network trading system by Andriy Sydoruk. It reads a trained model file, converts the model’s output into trading signals and combines those signals with order, equity and series controls. The developer identifies H1 as the preferred operating timeframe.

A trained model is required

The default configuration can place random orders. The developer instructs users to load a trained neural-network file and set Random Orders = false for model-based trading. Do this before evaluating the intended strategy. Attaching the EA without its model and configuration does not establish that it is running a trained system.

The listing links to the separate free Teaching T-INN training utility. Its instructions place the script in the Scripts folder and run it on the intended instrument and timeframe after loading its price history. The output model uses a Files location, with the trainer defaulting to the folder shared by MetaTrader terminals. Check the configured filename and location when connecting it to the EA.

The public discussion supplies an EURUSD H1 example from September 2022. The developer’s setup guide says to rename the supplied file’s extension from .sqlite to .net, place it in Terminal/Common/Files, match Net File Name and enable File Common Data On. That older example explains the workflow; its availability does not establish suitability for today’s market.

Signal sensitivity and trade management

The developer calls the architecture T-INN, combining clustered training data with neural-network processing. The practical trading control is Level Signal, which adjusts the response to the network’s output. The source also lists the following management areas:

• Position and exit settings: money management, stop loss, take profit, trailing start and trailing stop.
• Equity controls: TotalEquityLoss and TotalEquityProfit.
• Order limits and spacing: LimitOrders and GridStep.
• Series modes: TypeSeria, SeriaTrendOn and SeriaAntyTrendOn.
• Monitoring: configurable comments and the ability to write its own log files.

The grid-spacing and series inputs matter when assessing exposure. This is not presented as a no-grid system, and the model’s signal is only one part of the account risk. Review the order limit, sizing and equity settings together before enabling automated trading.

Separate model training from validation

Teaching T-INN uses agTeaching to train, agAudit to inspect the training interval and agForward to check the later forward period. Its guide also explains the cluster-count setting. A network can fit its training period well without performing similarly on later data: keep these intervals separate, inspect the trade sequence and then test the configured model on a demo account.

The November 27, 2022 release notes describe a fix for loaded networks remaining tied to a specified timeframe despite the chart or tester selection. Check the actual timeframe behavior when evaluating an existing model.

This EA is most relevant to traders willing to manage model files and test their assumptions. Public examples are separate from the purchase; no private developer consultation or newly trained model is promised here.

Intel Neuron Bot specifications · Developer’s public discussion.

Included: standalone MT5 delivery, account download instructions after payment, and Free lifetime updates. Follow your order instructions and confirm the required trained model before use.