Machine learning in trading: theory, models, practice and algo-trading - page 2023
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
English... it's closed access on the medium, you need a paid subscription
I'm making another bot right now... this one's just in testing...
Add lstm or gru and some backing... and then continue working on new NS architectures for time seriesMax, slow down, use your head...
1) The algorithm works only because of the intricate layers of the type ,
2) He's clearly a man with special education, and obviously not a fool.
3) His code is from 2020, which means that he knows for sure and touched lstm and gru, but did not apply them, and wrote his
So naive to think that you read two manuals and turn the world around) Each algorithm has its own problem and its own application!
So I would recommend you to slow down and think more logically, to better understand why it works, rather than chasing trendy algorithm names
Max, slow down your ardor, think with your head a little bit...
1) The algorithm works solely because of the intricate layers of the type,
2) He's obviously a man with special education, and he's not stupid.
3) His code is from 2020, which means that he knows for sure and touched lstm and gru, but did not apply them, and wrote his
So naive to think that you read two manuals and turn the world around) every algorithm has its own task and its own application!
So I would recommend to slow down and think more logically, to better understand why it works, rather than chasing trendy algorithm names
They use very old models, I've already written. Nothing super cool. I think that such labs are now written by students
no long memory, the model itself is slow
some deep q-learning will work just as well
You don't know what to say.) I haven't read 2 manuals, I've read several books on NS and reinforcement. I just didn't use grids.
They use very old models, I've already written. There's nothing super cool. I think this is the kind of lab that students write nowadays.
no long memory, the model itself is slowYes, the models are old, but he doesn't use the model as it is, he makes a neuron out of each model, and the neuron into layers...
And the old model was used as is, the model is old, but it's just a neuron in the new architecture.
As for memory, he just implemented long memory, and with a noise filter. Where he wrote "make coffee," where he wrote "to make coffee", that's exactly the memory that suits the market, I'm working on a similar problem myself ..... I doubt that modern "pop" architectures can do that, they have other tasks...
But try it, let's hope I'm wrong, if so it's much simpler, everyone will be better off for it.
Yes, the models are old, but it doesn't use the model as is, it makes each model into a neuron, and the neuron into layers...
And the old model was used as is, the model is old, but it's just a neuron in the new architecture.
Regarding memory, he just implemented a long memory, and with a noise filter. Where he wrote "make coffee," where he wrote "to make coffee", that's exactly the memory that suits the market, I'm working on a similar problem myself ..... I doubt that modern "pop" architectures can do that, they have other tasks...
But try it, let's hope I'm wrong, if so it's much simpler, everyone will only be better off.
about "making coffee" is a textbook example from rl books, about Markov chains :D The basics.
Read about problems of vanilla recurrent networks and how they are solved in lstm. Vanilla's are not capable of long memory, so don't argue.
About "making coffee" is a textbook example from rl books, about Markov chains :D Fundamentals.
Read about problems of vanilla recurrent networks and how they are solved in lstm etc. Vanilla's aren't capable of long memory, so don't argue.
Ok, I won't argue anymore))
Do it in gru , and you'll see the results.
Okay, no more))
Do it in gru , and then show the result
No one is saying that the architecture will be pop. Just more advanced bricks.
ready-made, in python mostly
Can you throw me a link)))?
https://pytorch.org/
If anyone has ideas for AI projects or other digital projects that could be really in demand and there are some customers/sales markets, it is possible to get a grant for startups up to 3 million rubles with the possibility to pay employees up to 100 thousand rubles per month.
As a result, an object of intellectual property (program, invention, etc.) must be registered
https://ит-гранты.рф/
Only on accountability, coordination, evaluation, business plan, etc. paper will have to spend a bunch, if not one)
Priority areas:
https://ит-гранты.рф/pnp