My 10,737,341,824 combinations will take 697,682 hours to complete
I've got an optimization I'd like to run, but the 10,737,341,824 combinations want to take 697,682 hours to complete. There's a very good chance, I'll be dead by then.
However, is there any way the processing can be distributed amongst 1,000 computers? If there is, I could complete this project in one month.
Thanks,
Vooch
However, is there any way the processing can be distributed amongst 1,000 computers? If there is, I could complete this project in one month.
Thanks,
Vooch
lol ...
Hopefully you live a few years and then you can buy at your food shop arround the corner a computer who calculates that in 10 mins.
I rember doing raytracing on a 486. For a 60 sec. scene, with a lot of light and mirror effects I used my italy vacation of 3 weeks. when I came back, I had a 20 MB avi file with a mindburstingly high resolution of 320x240. Now I play UT2004 with 1600x1200 and about 60 frames with about 100 times the objects, I had in this animation. Calcualted in realtime. Never worry, how long something will take :-)
On a related note: I had a DeMarker parameter analysis for the 3 pairs USDCHF/GBPUSD/USDEUR on 3 different computers for a nice overnighter. With frightening results .... How come DeMarker was SO right?
Vooch,
We all have the same problem....
U need to break up the variables into two or three sets, optimise 1 set, use the best parameters, optimise the second set, fix the parameters, optimise the third, fix the parameters...and then start all over again!
That way, it will only take u 1 day to do an iteration!
enjoy... lol
We all have the same problem....
U need to break up the variables into two or three sets, optimise 1 set, use the best parameters, optimise the second set, fix the parameters, optimise the third, fix the parameters...and then start all over again!
That way, it will only take u 1 day to do an iteration!
enjoy... lol
Vooch,
I know you want to be 70 years old before you discover the correct optimised settings :) but apart from hbd's suggestions, you could also look at having a much broader degree in your 'step' values. i.e. if you have 30 steps in the range in step values of 1, try changing the step rate to say 5 or more. This would at least narrow down the optimal range.
Once determined, narrow the Start and End values and fine tune the 'step' values to zero in. I'm sure the degree of success is dependent on the experts inner workings but this logic may help.
Cheers
Martin
I know you want to be 70 years old before you discover the correct optimised settings :) but apart from hbd's suggestions, you could also look at having a much broader degree in your 'step' values. i.e. if you have 30 steps in the range in step values of 1, try changing the step rate to say 5 or more. This would at least narrow down the optimal range.
Once determined, narrow the Start and End values and fine tune the 'step' values to zero in. I'm sure the degree of success is dependent on the experts inner workings but this logic may help.
Cheers
Martin
> I know you want to be 70 years old before you discover the correct optimised settings :) but
> apart from hbd's suggestions, you could also look at having a much broader degree in your
> 'step' values. i.e. if you have 30 steps in the range in step values of 1, try changing the step
> rate to say 5 or more. This would at least narrow down the optimal range.
hehehe... I've already got that optimized.
The start is 0, the step is 1, the stop is 1
30 external variables
They are all switches which are used to determine the best combination to use
I've been thinking about some alternate ways to approach the problem and get a solution.
Until then, only 697,676 hours to gooooo.....
hehe
> apart from hbd's suggestions, you could also look at having a much broader degree in your
> 'step' values. i.e. if you have 30 steps in the range in step values of 1, try changing the step
> rate to say 5 or more. This would at least narrow down the optimal range.
hehehe... I've already got that optimized.
The start is 0, the step is 1, the stop is 1
30 external variables
They are all switches which are used to determine the best combination to use
I've been thinking about some alternate ways to approach the problem and get a solution.
Until then, only 697,676 hours to gooooo.....
hehe
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
However, is there any way the processing can be distributed amongst 1,000 computers? If there is, I could complete this project in one month.
Thanks,
Vooch