New MetaTrader 5 Platform beta build 2245: DirectX functions
for 3D visualization in MQL5 and symbol settings in Strategy Tester
18. Tester: A plethora of new features and improvements: ...
read more here
Good article was published -
Continuous Walk-Forward Optimization (Part 1): Working with
In the previous articles (Optimization Management (Part I)
Optimization Management (Part 2)) we considered a
mechanism for launching the optimization in the terminal through a third-party process. This allows creating a certain Optimization
Manager which can implement the process similarly to a trading algorithm implementing a specific trading process, i.e. in a fully
automated mode without user interference. The idea is to create an algorithm which manages the sliding optimization process, in which
forward and historical periods are shifted by a preset interval and overlap each other.
This approach to algorithm optimization can serve as strategy robustness testing rather than pure optimization, although it performs
both roles. As a result, we can find out whether a trading system is stable and can determine optimal combinations of indicators for the
system. Since the described process can involve different robot coefficient filtration and optimal combination selection methods,
which we need to check in each of the time intervals (which can be multiple) the process can hardly be implemented manually. Moreover,
thus we can encounter errors connected with data transfer or other errors related to the human factor. Therefore, some tools are needed
that would manage the optimization process from the outside without our intervention. The created program meets the set goals. For a
more structured presentation, the program creation process has been split into several articles, each of which covers a specific area
of the program creation process.
This part is devoted to the creation of a toolkit for working with optimization reports, for importing them from the terminal, as well as
for filtering and sorting the obtained data. To provide a better presentation structure, we will use the *xml file format. Data from the
file can be read by both humans and programs. Moreover, data can be grouped in blocks inside the file and thus the required information
can be accessed faster and easier.
Our program is a third-party process written in C# and it needs to create and read created *xml documents similarly to MQL5 programs.
Therefore, the report creation block will be implemented as a DLL which can be used both in MQL5 and C# code. Thus, in order to develop an
MQL5 code, we will need a library. We will first describe the library creation process, while the next article will provide description
of the MQL5 code that works with the created library and generates optimization parameters. We will consider these parameters in the
Continuous Walk-Forward Optimization (Part 1): Working with Optimization Reports
Continuing with Part 2
Continuous Walk-Through Optimization (Part 2): Mechanism for creating an optimization report for any robot
This is the next article within a series devoted to the creation of an automated optimizer, which can perform walk-through optimization of trading strategies. The previous article described the creation of a DLL to be used in our auto optimizer and in Expert Advisors. This new part is entirely devoted to the MQL5 language. We will consider optimization report generation methods and the application of this functionality within your algorithms.
The strategy tester does not allow access to its data from an Expert Advisor while the provided results lack details, therefor,e we will use the optimization report downloading functionality implemented in my previous articles. Since separate parts of this functionality have been modified, while others were not fully covered in earlier articles, let's consider these features once again as they constitute the key parts of our program. Let's start with one of the new features: addition of custom commission. All classes and functions described in this article are located under the Include/History manager directory.
Forum on trading, automated trading systems and testing trading
New version of MetaTrader 5 build
2340 platform: Managing account settings in the tester and expanding integration with Python
Renat Fatkhullin , 2020/02/25 19:46
Forum on trading, automated trading systems
and testing trading strategies
How do I configure
MT5 to use ALL "Agent -> Local: 4 cores" during Strategy Tester ??
Agents are used during optimisation. For a backtest
only one agent is required
today I tried tests on my local farm and my metatrader 5 used on linux dissapeared my agents, I tried installing metatester alone but still
and journal states "2020.04.18 17:15:22.124 Tester Cloud servers switched off"
today I tried tests on my local farm and my metatrader 5 used on linux dissapeared my agents, I tried installing metatester alone but
still not work.
It may be some limitation ...I know that cloud does not work in VPS and in 32-bit Metatrader (but I am not sure about Linux ... it may
be same limitation):
I want to use different subset of PC in like IP 22.214.171.124 ,not use local network 192.168.1.5
192.168.1.5 is pass ok connect and work good.
126.96.36.199 log always show connect .... (no password issue or task going on..)
If this is possible to do?
I make more test case .. (And there is no debug message to make sure what happen ?! It's sock ><)
Test Env MT5 Build 2410(08 May 2020) / Win10 x64 base /All PR >120 All software use same version.
PC2 188.8.131.52 --->(192.168.18.5)
PC3 192.168.18.8 (Ubuntu)
Case A NB can see PC1 (But speed is limit by lowest one, it's look like load balance is not working? )
Case B PC1 can't see NB
Case C NB,PC1 can't see PC2
Case D PC1 can see PC1 in local net.
Case E NB can see PC3 (ubuntu after add winbind)
I try different way to use. And PC1 get several agents inside , I don't know if it will get side effect ?
And I try to check firewall ,and remove agents and add it again.
It's not work ><
Continuous Walk-Forward Optimization
"Due to the apparent lack of memory with an excessive number of agents and a decrease in the speed of calculations on hyper-threading cores, we decided to limit ourselves to only physical cores when working in cloud.
We have long been evaluating the approximate resource sufficiency of agents before issuing tasks to them, and one of the most effective is to work only on physical cores in cloud.
Locally, you can use all the cores as you can easily control their shutdown."
On my new hardware (AMD Ryzen 9300, 32GB DDR4) I am observing a number of agent results - that were (presumably) running on hyper-threaded cores, produce erroneous results in the strategy tester.
So, as this appears to me, it's not possible to use all the cores locally - or can anyone confirm that testing works on his/her hyper-threading cores?