Machine learning in trading: theory, models, practice and algo-trading - page 2314
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
Lopez De Prado's 2nd book is out, I liked the first one. The second promises to be no less interesting
How did you manage that?
And what is the output of what dimensions?
When forecasting the network, I pull out the states of the weights in the layers, the layers in the form of matrices, I make a dataset from the matrices and send it to "youmap". The output is 2 dimensions.
networks from the "neuralnet" packageWhen forecasting the network, I pull the states of weights in the layers, the layers as matrices, from the matrices I make a dataset and "yumap". The output is 2 dimensions.
networks from the "neuralnet" packageI see. So, what is the idea?
It makes sense to divide training data into parts and use non-intersecting parts when training each layer of the neural network. (Winwector's idea).Haven't tried it, but they claim it's useful.
Good luck
I see. What is the idea?
It makes sense to divide the training data into parts and use the non-intersecting parts when training each layer of the neural network. (Winwector's idea).Haven't tried it, but they claim it's useful.
Good luck
My idea was the following
1) to train neural network in some actions let it be bye/seat
2) the network on new data will make a lot of mistakes
3) I wanted to cluster the patterns in the layers of the network to see if I could separate the wrong decisions of the network from the right ones by looking at the patterns that occur in the network during signal processing...
=========
Vladimir, do you know if there is a package in R-ka where I can interact with the graphics, for example, so I can select an area on a graph with my mouse and get the parameters of this area in the code
This is a small real account on BitMEX.
Bot enters on public signals neuron and closes itself, full automatic.
Maximum pose of no more than 30% of the depo.
***
Now the initial easiest version, no stops, waiting for when it will sell everything))
Neuronet is learning to trade on the real market.
This is a small real account on BitMEX.
Bot enters on public signals neuron and closes himself, full automatic.
Maximum pose of no more than 30% of the depo.
***
Now the initial is the easiest version, no stops, waiting when it will sell out))
and where did the link to monitor the account go?
Apparently the bot removes the link. Judging by how fast.
Apparently the bot removes links. Judging by how quickly.
I know who this bot is))
New Catbust features
uncertainty prediction is an interesting thing, similar to Active Lerning, about which I wrote an article