# is there a way to export data as 3D-array so I can "navigate" on cells more efficiently?
**URL:** https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327
**Category:** ParaView Support
**Tags:** python
**Created:** [May 31, 2021, 9:15am UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327 "2021-05-31T09:15:22Z")
**Posts on this page:** 7
**Page:** 1
### Author:  [@coreman](https://discourse.paraview.org/u/coreman)
#### Post date: [May 31, 2021, 9:15am UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/1 "2021-05-31T09:15:22Z")
Hi there!
Basically I’m working with “pandas” and want to create an algorithm that:
Creates a group of Cells (with some criterion), then checks its “neighbouring” cells.
Thing is, I cant easily track “Neighbouring” cells when I save data as CSV. Point ID doesnt give me much info on how to move from a cell to another cell.
Has any of you guys have done something similar?
Know a way around it?
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [May 31, 2021, 1:05pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/2 "2021-05-31T13:05:12Z")
CSV format does not support to save cells. That being said, you could GenerateCellIds and then try to put the multiple ids on the points.
---
### Author:  [@coreman](https://discourse.paraview.org/u/coreman)
#### Post date: [May 31, 2021, 1:15pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/3 "2021-05-31T13:15:12Z")
Im sorry. I dont understand what u mean ☹
Point ID is being generated, and my lists are -Point ID, X,Y,Z, (other quantities- but I still cant go to neighbouring cells based on Point ID.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [May 31, 2021, 1:18pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/4 "2021-05-31T13:18:49Z")
Indeed, you need to write your own ProgrammableFilter that will add the multiple CellIds on each points.
That being said, I’m not sure to see the point of it.
---
### Author:  [@coreman](https://discourse.paraview.org/u/coreman)
#### Post date: [May 31, 2021, 1:33pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/5 "2021-05-31T13:33:14Z")
The point is that I want to examine a quantity in the neighbouring cells of a group of cells.
I want to do that via python so it will be easier to add a unity to an index and go to the next cell on the direction of the index.
I have no way of finding neighbouring cells, with the form Paraview is exporting the data.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [May 31, 2021, 1:34pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/6 "2021-05-31T13:34:52Z")
TBH you should not use .csv format for that. You can just use VTKPython and read a native vtk format file.
---
### Author:  [@coreman](https://discourse.paraview.org/u/coreman)
#### Post date: [May 31, 2021, 1:53pm UTC](https://discourse.paraview.org/t/is-there-a-way-to-export-data-as-3d-array-so-i-can-navigate-on-cells-more-efficiently/7327/7 "2021-05-31T13:53:35Z")
Thank you. will try it