Machine learning in trading: theory, models, practice and algo-trading - page 1392
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
Something seems to me that sockets with python tester will not work correctly. At least because of the inconsistency of the threads.
What threads? The python server is constantly running, the tester client pokes at it, gives and takes information, shuts down... it's the same as in the real world.
They're worried about security... ddos attacks on themselves or their servers... I dunno, I don't know...
Well, if you do it in the cloud, you can dostost sites from different ip numbers... but in the cloud to disable the possibility and allWell they're not such suckers these liquidity providers to give easy holes. They have a whole staff of geeks who can't tie their own shoelaces, but know all about machine learning mathematics and statistics. To beat them you really have to be that dickfx. The last dough machine broke down when the ISPs closed the hole. That's how you have to hold the market by the balls to change the movement of all currency pairs on a regular basis.
And she's so broken that she's writing articles here now.
Kindergarten.
Imho, streaming data science is more suitable for the market
I don't know anything about it, just came across it, but it seems to be useful
I've touched the subject of data streams and rough paths before, but still haven't understood it
https://towardsdatascience.com/why-you-should-learn-about-streaming-data-science-2f603ad24aea
What threads? The python server is constantly running, the tester client pokes at it, gives and takes information, shuts down... it's the same as in the real world.
They're worried about security... ddos attacks on themselves or their servers... I don't know, I don't know...
Well, if you do it in the cloud, you can ddos sites from different ip numbers... but in the cloud, just turn off the possibility and that's allImho, streaming data science is more suitable for the market
I don't know anything about it, just came across it, but it seems to be useful
I've touched the subject of data streams and rough paths before, but still haven't understood it
https://towardsdatascience.com/why-you-should-learn-about-streaming-data-science-2f603ad24aea
If I understand correctly, it looks like sequential analysis, optimal stopping tasks etc.
By the way, Berezovsky worked in this field when he was a mathematician.)
If I understand correctly, it is similar to sequential analysis, optimal stopping problems, etc.
By the way, Berezovsky worked in this field when he was a mathematician)
It looks like this, only with the use of MO.
Berezovsky had time to write esoteric books.)
Not only a week, but the efficiency decreases over time, which is natural.
I am testing on real ticks of the broker I trade with, so this is not a hope but an accurate test.
I read the articles, yours and not yours. I've never thought about it before. Imho, it is more complicated than in your articles. With RNN, I don't know. RNN is asking for it, but how and where to put it, I do not know. Once read the theory, but it all slipped out because it was useless).
In the Internet there are interesting articles on the TF and Keras, maybe from there you can pluck something).
I read the articles, yours and not yours. RL is an interesting direction, I had not thought of it before. Imho, it is more complicated than in your articles. With RNN, or something, I do not know. RNN is asking for it, but how and where to put it, I do not know. Once read the theory, but it all slipped out because it was useless).
There are some interesting tricks on the internet on TF and Keras, maybe you can pluck something from there).
Figure out the simple stuff first. Even experts with experience cannot make RL work in the market.
It's an interesting direction, but everyone drops out like a magician because he lacks the traction.
I don't know how to make RNN work... Keras and TF... No way.
first deal with a simple one. Even specialists with experience cannot make RL work in the market.
The direction is interesting, but everyone throws like a magician, because there is not enough traction
I know everything, I need RNN... Keras and TF do not work.
I will figure it out. I just started with Python with NS, and I do not care which ones. Maybe I won't be RL, it's just a matter of a few days.
The classic version won't work, unfortunately... it's also all about the features/targets.