What kind of spatial relations are the basis to make groups? For example, two lines that touch each other should belong to the same group?
What kind of spatial relations are the basis to make groups? For example, two lines that touch each other should belong to the same group?
yes i am grouping connected lines in one group. IN the image i had given the network sample which i have.For selection criterion i am using the feature_ID if the count of feature_ID in one grp is more than 40 then go for regrouping
If the number of touching lines is greater than 40, how do you want to subdivide the group?
As a first pass I would use the networkTopologyCalculator to assign each touching line a common _network_id. I would then use either the Aggregator or StatisticsCalculator to get the count of the number of features in each network.
If that number is greater than 40, you have a couple of choices for subdivision.
You can divide them into groups of 40 by counting the features
You can use a random sampler to separate them into groups
You can tile the data (possibly iteratively) so that each tile contains a maximum of 40 features.
If the number of touching lines is greater than 40, how do you want to subdivide the group?
As a first pass I would use the networkTopologyCalculator to assign each touching line a common _network_id. I would then use either the Aggregator or StatisticsCalculator to get the count of the number of features in each network.
If that number is greater than 40, you have a couple of choices for subdivision.
You can divide them into groups of 40 by counting the features
You can use a random sampler to separate them into groups
You can tile the data (possibly iteratively) so that each tile contains a maximum of 40 features.
Hi,
I have tried this approach also but i am not getting the results as per my need . So i had decided to of or simple approach of using buffer. In below WB i am using tester to test my condition and then feature which passes test condition >30 goes for regrouping . for regrouping i am using 2nd criterion as for grouping as i am creating the area of network and then using spatial relator to grp the feature. but things are not working in this case .
i so i want to use buffer and then something which will create small groups bases on my test conditions but not sure how i can do this.
If the number of touching lines is greater than 40, how do you want to subdivide the group?
As a first pass I would use the networkTopologyCalculator to assign each touching line a common _network_id. I would then use either the Aggregator or StatisticsCalculator to get the count of the number of features in each network.
If that number is greater than 40, you have a couple of choices for subdivision.
You can divide them into groups of 40 by counting the features
You can use a random sampler to separate them into groups
You can tile the data (possibly iteratively) so that each tile contains a maximum of 40 features.
HI Jdn ,
could you please elaborate more on subdivision part.
Thank you in advance.
What kind of spatial relations are the basis to make groups? For example, two lines that touch each other should belong to the same group?
Hi @takashi
In this case i am creating the buffer of 1 m around each road and then dissolving it . then i am giving number to each area using counter. Finally i am using line on area overlay get the line in each buffer area. these line area getting number from these area. now i want to check if the number of line in each area. if that is more than 30 then renaming line in that group should should get different group ID. it is very important to group line bases on there spatial location . that is why i am using simplest method go buffer. please suggest how i can go for sudivision in each group.?Thank you in advance.
Hi,
@ jdh and @takashi i am still not finding the right way to do create sub groups in groups .