Machine learning in trading: theory, models, practice and algo-trading - page 479
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
Well, then I, alas, can not help - the problem must be reproduced before you solve it.
For a delay you can try to useEventSetTimer - read the indicator in a couple of seconds after the bar appears.
Thanks for the recommendation.
Can you do it not in a couple of seconds, but in about 30 seconds? How to do it?
Can I do it in 30 seconds instead of a couple of seconds? How do you do that?
You can. Look for indicators in the code base on his example.
On my question, I understand, there are no specialists????
Does anyone use APi from Nvidea. Just looked around the other day in terms of cards, I see that the company produces cards for mining, without video connectors.
Does anyone use APi from Nvidea. Just the other day I looked around in terms of cards, I see that the company produces cards for mining, without video connectors.
I forgot, and some API is released, i.e. CUDA
There are a lot of nvidia api -
OpenCL is a popular standard, you need to write a special C code. It is supported by both Nvidia and AMD, you can even install the library for cpu processors instead of the video card. I use in R through the libraryhttps://cran.r-project.org/web/packages/OpenCL/index.html.
MT5 can do this api too.
CUDA is some kind of strictly nvidia library, and for example AMD vises do not support it. The point is to take regular C++ code and compile it so that it runs on the graphics card. Perhaps there are restrictions on permissible functions, I haven't got into details.
CUDNN - api and bibliotec of the CUDA-based neural network
There is such a neuron MXNET, it supports all these technologies, you can use it to test the videohttps://mxnet.incubator.apache.org/get_started/install.html
Help me understand the process :)
On the example of a linear classifier. Suppose we teach something to the classifier, we feed two outputs with price increments from 0 to 1, where 0.5 is no signal (sum of outputs always equals 1)
If we simply draw a regression line, for example, according to the prices, the larger price increments will lie further from the line, the smaller ones closer, i.e. if the classifier outputs show 0.9; 0.1, the positive increment lies further from the line, i.e. the signal 0.9 will be stronger than the signal 0.6 to buy
And if we take a neural network with non-linear classification, will the outputs show the signal strength or will they only show the degree of belonging to 1 of 2 classes and no more
I.e., will this condition be met:
It seems to me, that on this situation already a half of the beginners, badly knowing the matrix, will fail... Because the degree of belonging to this or that class, intuitively, will talk about more / less signal strength. But is it really so and isn't it better to create more classes on which to distribute values of increments, say, in %, then falling of value into one or another class (one of 10, let's say) will already precisely tell the value of increment.Help me understand the process :)
On the example of a linear classifier. Suppose we teach something to the classifier and supply price increments from 0 to 1 to its outputs, where 0.5 is not a signal.
If we simply draw a regression line, for example, by prices, then larger price increments will lie further away from the line, smaller ones closer to it. That is if the classifier outputs show 0.9; 0.1, the positive increment lies far away from the line, i.e. the signal 0.9 will be stronger than the signal 0.6 to buy.
And if we take a neural network with non-linear classification, will the outputs show the signal strength or will they only show the degree of belonging to 1 of 2 classes and no more
That is, whether such a condition will be met:
Understand: the line is the boundary (in the simplest case, the average value). The closer to the boundary the increments of prices, the less they differ from the average, although in absolute value they may not be small at all. (for this formulation of the problem)
Think about it: the line is a boundary (in the simplest case, the average value). The closer to the boundary the incremental prices are, the less they differ from the average, although in absolute value they may not be small at all. (for this formulation of the problem)
That is, according to the degree of belonging of the target to the class cannot be concluded about the absolute change, 0.9 will not mean that the price increment will be greater than in the case of 0.6
That is, the degree to which the target belongs to a class cannot be used to infer an absolute change, 0.9 will not mean that the price increment will be greater than in the case of 0.6
Again, it depends on how the classification is constructed. In the example above, the classification was based on the distance from the center line (boundary) without taking into account the absolute value of the increment. If you introduce the absolute value of increment, the classification will be different in principle. Its scale will also be different.