Machine learning in trading: theory, models, practice and algo-trading - page 484
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
51 neurons. Configuration -15,20,15,10, 5,1. Yes, BP training with simulated annealing.
You definitely saw the results earlier. I can repeat, if anyone has a desire to see.
In general, if even once every 3 months to train, then 23 hours - yuck, compared to scratching a turnip with the usual TC on the logic.
I have a slow computer, on a modern one it will be at least two times faster.
If I look at the simulation I'll see if the real inputs are correct (I'll have more entries later), training time is 2 seconds (between last training and the penultimate one, I use one more time in the queue just for fun, to see if it eats memory)
And it seems to me this is slow, strongly not to copy the hyperparameters. If the finished model fits in a minute or two on this set, it will be bearable... I would be crazy to wait 23 hours :)20 000 bars set, 2 inputs 1 output, 200 trees (just testing, then I will add more inputs), training time of 2 seconds (between the last training and penultimate, I train a lot of times in a row for fun, to see if it eats memory)
And it seems to me this is slow, strongly not to copy hyperparameters. If the finished model fits in a minute or two on this set, it will be bearable... I would be crazy to wait 23 hours :)Respectfully.
You can opt for this speed online, adjusting parameters to the specific situation.
Sincerely.
That's what I'm talking about... this model is not parallelized yet, I'm sure there are packages where it flies at all
That's what I'm talking about... this model is not parallelized yet, I'm sure there are packages where it flies at all
Sincerely.
I've paralleled calculations via dll in c++ when there was no mt5 yet, but since I switched to mt5 I've forgotten what it is.
Sincerely.
alglieb on pluses and sharpe have all the libs parallelized + even with support for video cards in my opinion, but they are already paid
alglieb on pluses and sharpe have all the libs parallelized + even with video card support in my opinion, but they are already paid
P.S. There's nothing complicated there, look at the number of cores (threads), create as many threads in your program as cores are available and put needed data and parameters into each of them, create a queue of calculations for each thread and put new settings there, CPU load is 100%.Regards.
Sensei))))) Well, it's not my fault I have such teachers. When decoded, the message looks like this
Teacher! As long as you're jerking off in parrots .............mat.........mat..........selected russian mat.............mat,
which can't even be described by dots... put R! Collect the rattle devoid of previously voiced features, because now you
know everything, including blending, stacking. Stack Misha, stack)))
Hilarious... hilarious!!!!! Just took my breath away.... Can't you fix the code file? Not just to make a fuss, but to do something???
He no longer gives out errors, but the line does not want to output :-(
hilarious)))
Why all of a sudden? Explain!!!!
I'm now experimenting to go into pure MO, without the old forex habits. No plotting profit charts to evaluate the pattern, no wands or other indicators. Instead, regression (gains per bar ahead), complex crossvalidations and special patterns. I managed on eurusd m5 to train the model on 10000 bars of history, get r^2 of about 0.001, or almost 52% accuracy, and stay with the non-deteriorating result in the future. At zero spread it looks good, even a spread of 2 five digit points would pull in profits. But that's not enough.
There is also a strong fear that dealing centers have seriously started to destroy working Advisors and this entire search for the universal grail makes no sense. There is one dealing company that has been destroying working EAs for a couple of years already and even if the EA has been bringing profit for months, all of them have lost the balance within a week. This time there are more and more strange dealers, and I have only one dealer left, whom I trust. These are troubled times to come.
Why not open an account in a normal market? There are instruments in which one plus one tick is already a profit, no spreads, delays, candlestick drawing, fake-outs, brakes, etc...