The utility is designed to increase the efficiency of manual trading and develop auto-trading methods using neural networks without the need to resort to third-party programs, resorting to the import/export of working data. The forward period will allow you not only to test the strategy on time areas that are not available for optimization, but also to stop trading automatically under unfavorable conditions.

The genetic algorithm will search for the best parameters of stop loss, take profit, trailing stop, for 20 different strategies, or for each one separately.

It includes a trading panel for quick work in manual mode without distracting from working with the program. The new method of placing pending orders will allow you to place orders in two clicks, focusing only on the price chart, without paying attention to the long market numbers. The ability to save templates will allow you to quickly assess the market situation without leaving the program.

Visual evaluation of indicators from the market has never been so fast. The method for quick visual evaluation of indicators allows you to change the indicator parameters using the slider, without having to call the indicator property.

• Selection

[spoiler title=”Read More…”]

Training and testing of a neural network is divided into three stages: training, testing, and trading.

Training – Network training.

Test-checks the network after each training epoch (if the error is less than the previous one, the network is saved).

Trade-here the network’s responses to data that the network did not see during training and testing are shown.

• Indicators

Any indicators normalized to values from -1 to 1 can be fed to the network. And they should never go beyond the extremes of values during training in the future.

You can create a set of up to one hundred indicators.

• Teacher

As a training set, you can choose any indicator, a built-in zig-zag, or the output result of another network.

• Training

Before you start training, you need to decide on the type of network, each of which has its own special qualities.

In the simplest language

MLP – A network with back propagation, has good predictive qualities.

GRNN is a network with absolute memory.

SOM-Splits it into a specified number of classes.

• Trading Panel

You need to allow auto trading for the panel to work. To set the price of a pending order, simply click on the price chart and click on the desired order.

• Optimization

Before starting optimization, you need to select the strategies that will be used for optimization.

Stop Loss, Take Profit, Trailing Stop, and trading signal levels are available for optimization.

• Visual evaluation of indicators

The parameters of the indicators can be changed using the slider. The indicator will be recalculated immediately.

Detailed description

[/spoiler]