New set files. Have been tested since 1.01.2016, open prices, m30, 5 digids.

Excuse my ignorance, but I do not understand the video. As you do bactesting, and you make it profitable.

You can explain to me, in detail how to perform the optimization, and how long it lasts. That is, every time you recommend the optimization.

Here is a link to the video obout optimization, in english with subtitles

https://www.youtube.com/watch?v=LX4uG3Jy58k

I have two questions. This EA even though it has "neurons", it does not self learn from when it operates in area. True? It's always based on previous learning, but that's just drawing a perfect backtesting. That is one of my doubts, if self learn as if they do other EA

On another side, once trained, I do not know if I have to pass the CSV I make from my home computer with TDS 2, I do not know if I have to copy it also in the VPS in the MT4 in the appropriate folder. Is that so?

EA I really like it, maybe rent it to try, but I'd rather read your comments about my doubts.

Ok thank you very much, I rent it now and if in real it comes close to the results, I will buy and appreciate. I am very surprised this EA.

I rent your EA yesterday and I have some doubts about calling them that way.

The EA consumes an awful CPU in the VPS, I have three currencies and it has risen from 15% to 65% and I have 4 nuclos at 2ghz and 8gbram, if it is profitable it is not bad.

The other question. When optimizing for bars is very good, it is incredibly fast, for ticks it is IMPOSSIBLE almost, it is very slow, it can take 3 hours to make a pass in the period of 8 months. This is a huge problem, which I tell.

Every day that passes, this is the SWAP commission, that happens every night. I do all this with TDS 2 (tick data suite 2), which aside from 99% ticks, I can do dynamic spreads, slippages, I add the price per lot and also calculates the Swap for each night.

If an operation is several days open, and assuming it is a small lot in EurUSD, say commission 0.35 and each night the Swap is 0.33, so that each day that passes commissions are added.

When calculating only by bars and without the TDS, I can not calculate more or less the final gain.

Experience has taught me that many EA's are profitable at 90% ticks, then they are not at 99% and more when you put commissions.

I have it in real but in tests and well, they are few hours and it goes well, but my doubt is when there are operations of many days.

I use 3 EA. They load CPU between 5 and15%.

It was made to work with open prices. In this way speed of optimisation is maximum. 

It holds deals days, so you need to think about comission and swap. In my experiense profit is more.

Hi Vladimir,

I've recently purchased your neural network. I have some trouble with the optimization process. I've watch your video tutorial but I dont understand how to optimize the stop loss, breakeven.. there are some months in the backtesting that the expert is really profitable, but others with similar parameters, the expert breaks the account. So, what I have to do for optimize with a control of the risk using stop loss, breakeven, amount of arrows for a signal..?

Regards.

Dear Author , would you please consider to crate a signal to show this EA in real account performance ? Thank you