| Pages:
1
..
29
30
31
32
33
..
54 |
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
Many thanks Peter, Would it be possible for you to document, the stats export and how this works with the added option? so i can solve my stats
export issues ?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by Daniel UK1  | | Many thanks Peter, Would it be possible for you to document, the stats export and how this works with the added option? so i can solve my stats
export issues ? |
yes, but too busy on other things right now.
you will want custom export expanded mode
the rest you could leave as default
im not certain it works, but file name, data streams should only write on the first run of AU. ie in AU when index on the left = 1 only
Thanks received (1):
+1 Daniel UK1 at 2020-07-29 12:38:26
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
next build we have the option to
1) force a min of x indicators if there are not enough green
default will be 5 {force min false}
2) force x indicators regardless of green
this is shown in screen shot
we will get 10 indicators {force min = true}
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
Sounds great Peter, does the setting in this view for green and orange have any impact of the behavior of i set to min 5indicators ?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
in next build, ability to force x indicators in 2 pass, ability to have a min of y indicators
count in stats of how many indicators are used
tertiary filters are now working (there were bugs) but disabled till we have ts code in gsb
and the stats. Progress on TF is happening, but extremely slow
im doing the tests on forced 10,8,6,4 indicators.
10 and 8 are done, but results not compiled from csv into excel
If you have a high end machine, your welcome to send me the share key to add the cloud used for this
My guess is processing done by Tuesday. Contributors get the research results.
|
|
|
OUrocketman
Junior Member
Posts: 18
Registered: 10-5-2018
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | in next build, ability to force x indicators in 2 pass, ability to have a min of y indicators
count in stats of how many indicators are used
tertiary filters are now working (there were bugs) but disabled till we have ts code in gsb
and the stats. Progress on TF is happening, but extremely slow
im doing the tests on forced 10,8,6,4 indicators.
10 and 8 are done, but results not compiled from csv into excel
If you have a high end machine, your welcome to send me the share key to add the cloud used for this
My guess is processing done by Tuesday. Contributors get the research results.
|
Peter, I've emailed you the share key which should net you 7 additional resources. Let me know if it's not cooperating.
Best Wishes!
Thanks received (1):
+1 admin at 2020-08-08 02:32:19
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
58.71 under testing
Note the default of 4 for min number of indicators.
IF you need to have indentical results on a benchmark, you may want to change this value to 1
However I expect 4 to be better
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
there is also settings pearsonspf so you can change the training pearsons / pf by macro
Daniel said he found lower pearsons / pf on indicator testing and higher on system build worked better
I tried lower on indicator testing and system build, and results were worse.
This is only for the adventurous users to play with
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
we also have custom export expanded
In this last build , for 1000 separator removed and "" as separator removed in custom export expanded
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | there is also settings pearsonspf so you can change the training pearsons / pf by macro
Daniel said he found lower pearsons / pf on indicator testing and higher on system build worked better
I tried lower on indicator testing and system build, and results were worse.
This is only for the adventurous users to play with |
Hi Peter,
Do you select the MAcros from C:\GSB\GSBAutomation\macrostoaddIntoGSB to insert the newer filters? Just making sure I do this to the right macros so
the AU picks them up. Thanks.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Bruce
edit the macros in gsb called green etc and the greenorange etc
im not sure what your trying to do.
the new gsb by default will have min of 4 in the macro
Note in your build macro, you want to add tellautomationthatgsbFinished
It makes automation think gsb is finished, which means stats doesnt slow the whole process down
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | Bruce
edit the macros in gsb called green etc and the greenorange etc
im not sure what your trying to do.
the new gsb by default will have min of 4 in the macro
Note in your build macro, you want to add tellautomationthatgsbFinished
It makes automation think gsb is finished, which means stats doesnt slow the whole process down |
I'm trying to do this;
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Bruce | | Quote: Originally posted by admin | Bruce
edit the macros in gsb called green etc and the greenorange etc
im not sure what your trying to do.
the new gsb by default will have min of 4 in the macro
Note in your build macro, you want to add tellautomationthatgsbFinished
It makes automation think gsb is finished, which means stats doesnt slow the whole process down |
I'm trying to do this;
|
Hi Bruce, are you saying it doesnt work? Im not clear on what you say
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
gsb counter trendhl and are faulty.
These will be renamed to counter trendhlv2, counter trendhl2v2
New eld will be supplied. Likely in the next build
Thanks to user Joe for spotting this.
|
|
|
meldinman
Junior Member
Posts: 45
Registered: 22-6-2020
Member Is Offline
|
|
I believe GSB_Moonsaw may be faulty as well.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
can you explain more?
I think the range of parameters supplied was the issue, but its been a long time since I looked at it
|
|
|
meldinman
Junior Member
Posts: 45
Registered: 22-6-2020
Member Is Offline
|
|
If I recall, the function was looking for only a weight parameter even though systems where building the code with a source varieable + a weight
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by meldinman | | If I recall, the function was looking for only a weight parameter even though systems where building the code with a source varieable + a weight
|
I dont see how that can compile. Next time you see it, can we do teamviewer.com?
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by meldinman | | If I recall, the function was looking for only a weight parameter even though systems where building the code with a source varieable + a weight
|
I dont see how that can compile. Next time you see it, can we do teamviewer.com? |
Hi Peter,
Maybe meldinman means the TS function only has a weight and no other variable, but GSB uses the moon function with a weight and a other variable?
But that is not the case, the TS function and GSB use variable "shift".
|
|
|
meldinman
Junior Member
Posts: 45
Registered: 22-6-2020
Member Is Offline
|
|
something like that carl. It wasnt compiling either way. Will take a screenshot next time .
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Tertiary filter stats has been done, but apply to settings doesnt work, and there is no TS/MC/NT code yet
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Tertiary filter stats has been done, but apply to settings doesnt yet work, and there is no TS/MC/NT code yet.
No update yet on when I get the TS code etc, but im as keen as anyone for it to happen
Thanks received (1):
+1 moresi522 at 2020-08-28 07:48:20
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
58.74 released. It has tf stats but I think its disabled for users except me. This build just allows me to do some TF testing on what cloud workers
are free.
I will release a build with tf and stats, but no ts code soon.
TF will only work for beta tester mode.
Big picuture is,
1) how do I use tf and stats
2) what tf do I add into GSB first
Im thinking of all the Andrea Unger type of pattern filters (daily bars)
ie trade if last daily bar is inside bar, (or the opposite etc)
Im open to ideas on this. Long term lots of stuff will be added, beta testers can see what works,
then we release into build for all users the stuff that works
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
| Quote: Originally posted by admin | 58.74 released. It has tf stats but I think its disabled for users except me. This build just allows me to do some TF testing on what cloud workers
are free.
I will release a build with tf and stats, but no ts code soon.
TF will only work for beta tester mode.
Big picuture is,
1) how do I use tf and stats
2) what tf do I add into GSB first
Im thinking of all the Andrea Unger type of pattern filters (daily bars)
ie trade if last daily bar is inside bar, (or the opposite etc)
Im open to ideas on this. Long term lots of stuff will be added, beta testers can see what works,
then we release into build for all users the stuff that works
|
Great progress..
1, I assume same as secondary filters and primary right? or do you see any difference here?
One thing to look out for i believe is to monitor the correlation ie the edge that the tf is having on the system... so perhaps an easy way in code to
switch filters on and off, to test going ahead after launching live systems or systems that is in test "bench" before production.
2, Andrea unger types of filters yes, you could also use n bars up then n bars down.. or reverse.. you can also add *this last one i would like to
say "i believe you should add" closing range of yesterdays bar with definable closing range ie top n % or bottom n %.. defined for long or short.
One has to be careful though with adding to many filters though, easy to "filter" oneself out to perfection during training 
Cheers
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Agreed Daniel. I like that range filter too.
If the TF shows all 250 systems to improve with test repeated 4 times, then I would be happy with that as an outcome
|
|
|
| Pages:
1
..
29
30
31
32
33
..
54 |
![[*]](images/davis/default_icon.gif){kind=icon}
