THE IDEA EXCHANGE - page 36
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Thank you! Interesting idea!
I have it showing on all TFs. But something is mixed up in the yellow fractal. On all TFs it is drawing at different distances and not the same as on the blue one.
Please, check it!
I have found the parameter that allows me to see yellow fractals without distorting distances on all TFs.
So, I must correct the code before this expression to avoid this nonsense! :))
Sometimes a "creative" crisis sets in and you start scouring the literature and other people's work in search of a fresh or well-forgotten idea. Let's create something like an "Idea Bank" and in this thread will share ideas about trading / autotrading. Ideas are different, new or not very good for some, original or trite, but a new interpretation, at first glance absurdly insane, ideas that we use or that we do not fit, tried or not get to that, implemented or waiting for their time, in short - all sorts. Who doesn't feel sorry for what.
To facilitate reading, the main idea underlined in blue, discussion, comments and other impulses of the soul ordinary. Join in.
In this vein, I propose to consider an EA that on poundbucks with native settings on M5 has somehow cleverly taken 100 pips in the last 23 hours, first on the sell - 50 pips, then on the buy - 50 pips.
/* decompiled deleted */
The topic has somehow faded away...
I was thinking a month ago about a robot statistician.
The idea is the following: it collects data on candlesticks, total length, shadow length relative to the body, proportionally and in percentages, etc. Then it receives data on the subsequent movement and, accordingly, from the collected data it further determines which position to open based on the current candlestick. The advantage of the robot is that it can be a scalper, as well as an interleaver and a medium-term, depending on the timeframe in which it is set up.
I have never got a chance to try it, but I have not received a chance to try it yet.) But I can't get my hands on it...
The contest uses a Kohonen layer of 250 neurons. We have to do about 1500 of them. It will take at least 100machine hours to train the neuronics, maybe more. Could be faster, if the training algorithm is different (faster). For 250 neurons it would take 10 hours. But the problem starts with training not the grid, but the Expert Advisor. Here is the problem. For my EA it took me three weeks, but this is to put it crudely. In fact, something was changing all the time. The net time is eight or ten hours. If I have 1500 neurons, it'll take me about 80-120 machine hours to train the Expert Advisor. But this is only for one currency. And we need to do as many of them as possible. I don't have enough resources for that. Even if I change computers.
I recommend reading F. Wasserman Neurocomputing: Theory and Practice. It is very well written. I can send it to you by email if you need it. I can and not only this one, but other literature as well.
Are there math neuron functions in mql4? Or should swith be used: case 1: input neuron......case 5: output neuron, and the weight is continue;
That's the way arrays are used.
I recommend reading F. Wasserman Neurocomputing: Theory and Practice. It is very well written. I can send it to you by email if you need it. Not only that, but other books as well.
Can you send it to me? Very interesting to read! As a fellow countryman to a fellow countryman:)
Can you send it to me? Very interesting to read! As a fellow countryman to a fellow countryman:)
Where should I send it?
Where should I send it?
Why so complicated, you can send a private message ))))
Or on RapidShare.