# How to randomly subset X% of selected points?

What methods are available in ArcGIS 10.2 to randomly subset a selection of points. For example, in the attached screenshot I am interested in keeping 20% of the selected points and deleting the rest. • Well I don't think there is a default method for selecting random points from layer. Did you try with python script? Or add-in? – Marcin D Nov 21 '13 at 20:08

Here's a python function that will select random features in a layer based on percent, ignoring current selection:

``````def SelectRandomByPercent (layer, percent):
#layer variable is the layer name in TOC
#percent is percent as whole number  (0-100)
if percent > 100:
print "percent is greater than 100"
return
if percent < 0:
print "percent is less than zero"
return
import random
fc = arcpy.Describe (layer).catalogPath
featureCount = float (arcpy.GetCount_management (fc).getOutput (0))
count = int (featureCount * float (percent) / float (100))
if not count:
arcpy.SelectLayerByAttribute_management (layer, "CLEAR_SELECTION")
return
oids = [oid for oid, in arcpy.da.SearchCursor (fc, "OID@")]
oidFldName = arcpy.Describe (layer).OIDFieldName
path = arcpy.Describe (layer).path
randOids = random.sample (oids, count)
oidsStr = ", ".join (map (str, randOids))
sql = "{0} IN ({1})".format (delimOidFld, oidsStr)
arcpy.SelectLayerByAttribute_management (layer, "", sql)
``````

Copy/paste this into the python shell in ArcMap.

Then in the shell type `SelectRandomByPercent ("layer", num)`, where `layer` is the name of your layer, and `num` is a whole number of your percent. A variation to find a subset selection as asked:

``````def SelectRandomByPercent (layer, percent):
#layer variable is the layer name in TOC
#percent is percent as whole number  (0-100)
if percent > 100:
print "percent is greater than 100"
return
if percent < 0:
print "percent is less than zero"
return
import random
featureCount = float (arcpy.GetCount_management (layer).getOutput (0))
count = int (featureCount * float (percent) / float (100))
if not count:
arcpy.SelectLayerByAttribute_management (layer, "CLEAR_SELECTION")
return
oids = [oid for oid, in arcpy.da.SearchCursor (layer, "OID@")]
oidFldName = arcpy.Describe (layer).OIDFieldName
path = arcpy.Describe (layer).path
randOids = random.sample (oids, count)
oidsStr = ", ".join (map (str, randOids))
sql = "{0} IN ({1})".format (delimOidFld, oidsStr)
arcpy.SelectLayerByAttribute_management (layer, "", sql)
``````

Finally, one more variation to select a layer by a count, instead of a percent:

``````def SelectRandomByCount (layer, count):
import random
layerCount = int (arcpy.GetCount_management (layer).getOutput (0))
if layerCount < count:
print "input count is greater than layer count"
return
oids = [oid for oid, in arcpy.da.SearchCursor (layer, "OID@")]
oidFldName = arcpy.Describe (layer).OIDFieldName
path = arcpy.Describe (layer).path
randOids = random.sample (oids, count)
oidsStr = ", ".join (map (str, randOids))
sql = "{0} IN ({1})".format (delimOidFld, oidsStr)
arcpy.SelectLayerByAttribute_management (layer, "", sql)
``````
• Nice use of `random.sample()`. – Aaron Oct 26 '15 at 21:26
• Thanks @Aaron. I updated the answer for a subset selection without exporting first. – Emil Brundage Oct 26 '15 at 21:33
• +1. Are there any known limitations on string length for the `sql` parameter? – Paul Oct 28 '15 at 19:01
• @Paul I just tested this code to select 100% of features with a layer that has nearly 4 million features, which resulted in a memory error. So while there doesn't appear to be a hard string limit, there is a dependency on memory. There is also an SQL item limit for Oracle SDE databases, which I've blogged about here: emilsarcpython.blogspot.com/2015/10/… – Emil Brundage Oct 28 '15 at 19:10
• Esri used this code in a blog support.esri.com/en/technical-article/000013141 – Emil Brundage Jul 16 '18 at 15:24

Generally, I also recommend using the spatial ecology tools as discussed by blah238.

However, another method you could try would be to add an attribute called Random to store a random number: Then, using the field calculator on that attribute, with the Python Parser, use the following codeblock:

``````import random
def rand():
return random.random()
``````

See image below:

This will create random values between 0 and 1. Then, if you want to select 20% of the features, you could select features where the Random value is less than 0.2. Of course, this will work better with many features. I created a feature class with only 7 features as a test and there were no values less than 0.2. However, it looks like you have plenty of features, so that shouldn't matter. • this method will return on average 20% of the features, which in some cases would be preferred. But if you want 20% every time, you can do as suggested, then sort the features by the random value and select the first 20%. – Llaves Nov 22 '13 at 4:04
• Esri used this process in a blog: support.esri.com/en/technical-article/000013141 – Emil Brundage Jul 16 '18 at 17:44

There is also an earlier Select features at random script from @StephenLead available for ArcGIS Desktop. Although written, I think, for ArcGIS 9.x, and last modified in 2008, I used it in about 2010 at 10.0, and it still worked well.

You could try Hawth's Tools: http://www.spatialecology.com/htools/rndsel.php

Note that the existing selection is not honored so you would have to make a feature layer from the existing selection first.

• Unfortunately, that version isn't compatible with ArcGIS 9.3 and above. Now it's called Geospatial Modelling Environment: spatialecology.com/gme – kenbuja Nov 21 '13 at 21:33
• Good point, here is the equivalent command in GME: spatialecology.com/gme/rsample.htm – blah238 Nov 21 '13 at 21:43
• The GME toolset does not work "within" ArcGIS, rather it is a stand alone tool – Ryan Garnett Sep 16 '15 at 20:08

Here's another random selection add-in for ArcGIS 10, the Sampling Design Tool. It will let you select 20% of the features in your dataset. However, this doesn't use a selected set to make a random selection, similar to the restrictions of the Hawth's Tools mentioned by blah238.

You could also use the Subset Features tool. According to the documentation:

Divides the original dataset into two parts: one part to be used to model the spatial structure and produce a surface, the other to be used to compare and validate the output surface.

One disadvantage is that you need the Geostatistical Analyst extension.