# Programmable Source: Vorticity and gradient from numpy data
**URL:** https://discourse.paraview.org/t/programmable-source-vorticity-and-gradient-from-numpy-data/3846
**Category:** ParaView Support
**Tags:** python
**Created:** [March 18, 2020, 10:06pm UTC](https://discourse.paraview.org/t/programmable-source-vorticity-and-gradient-from-numpy-data/3846 "2020-03-18T22:06:59Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@MiguelAngelAGUIRRE](https://discourse.paraview.org/u/MiguelAngelAGUIRRE)
#### Post date: [March 18, 2020, 10:06pm UTC](https://discourse.paraview.org/t/programmable-source-vorticity-and-gradient-from-numpy-data/3846/1 "2020-03-18T22:06:59Z")
Hello,
I have a programmable source that creates a Delaunay2D mesh from Numpy arrays. Now I want to calculate the vorticity from the vector field but “algs.vorticity(Vector\_Speed)” do not work because “Vector\_Speed” is a numpy object. The same happens with the “algs.gradient(Vx)” (because “Vx” is a numpy array)
Since I really want to use the “algs.vorticity” function, I tried to convert the numpy vector (i.e., “Vector\_Speed”) into a vtk object and then to wrap this element but I cannot make it work. Please, can you provide me some guidelines to solve this issue?
Thanks in advance!
PS: the programmable source must be a vtkPolyData (vtkImage is not an option for my application)
Miguel
```
# MODULES LOAD:
import numpy as np
import vtk
from vtk.numpy_interface import algorithms as algs
from vtk.numpy_interface import dataset_adapter as dsa
import paraview.vtk.util.numpy_support as vnp
# NUMPY INPUT DATA:
X=np.array([0,1,2,0,1,2]) # X-coordinate
Y=np.array([0,0,0,1,1,1])
Z=np.array([0,0,0,0,0,0])
Vx=np.array([0,1,2,0,1,2]) #X-velocity
Vy=np.array([0,0,0,0,0,0])
Vz=np.array([0,0,0,0,0,0])
# DELAUNAY PLANE CREATION:
points = vtk.vtkPoints()
for i in range(X.size):
points.InsertNextPoint(X[i],Y[i],Z[i])
aPolyData = vtk.vtkPolyData()
aPolyData.SetPoints(points)
aCellArray = vtk.vtkCellArray()
boundary = vtk.vtkPolyData()
boundary.SetPoints(aPolyData.GetPoints())
boundary.SetPolys(aCellArray)
delaunay = vtk.vtkDelaunay2D()
if vtk.VTK_MAJOR_VERSION <= 5:
delaunay.SetInput(aPolyData.GetOutput())
delaunay.SetSource(boundary)
else:
delaunay.SetInputData(aPolyData)
delaunay.SetSourceData(boundary)
delaunay.Update()
self.GetOutput().ShallowCopy(delaunay.GetOutput())
# GRADIENT AND VORTICITY:
Vector_Speed = algs.make_vector(Vx,Vy,Vz)
#VORTICITY = algs.vorticity(Vector_Speed) #This fuction is not working here because Vector_Speed is a numpy element
#CONVERTING THE NUMPY VECTOR INTO A VTK ARRAY
vtkdata = vnp.numpy_to_vtk(Vector_Speed.ravel(), deep=True, array_type=vtk.VTK_FLOAT)
vtkdata.SetNumberOfComponents(3)
vtkdata.SetName("Vector_Speed")
#HERE I TRY TO WRAP THE VTK ELEMENT (but fails!)
WrapData = dsa.WrapDataObject(vtkdata.GetOutput())
Vector_Speed = WrapData.PointData["Vector_Speed"]
# THIS IS WHAT I WANT TO CALCULATE
VORTICITY = algs.vorticity(Vector_Speed)
Vx_GRADIENT = algs.gradient(Vx)
```
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [March 19, 2020, 1:40am UTC](https://discourse.paraview.org/t/programmable-source-vorticity-and-gradient-from-numpy-data/3846/2 "2020-03-19T01:40:00Z")
> **[Improved VTK - numpy integration - Kitware Blog](https://blog.kitware.com/improved-vtk-numpy-integration/)**
>
> Recently, we introduced a new Python module called numpy\_interface to VTK. The main objective of this module is to make it easier to interface VTK ... Read More
---
### Author:  [@MiguelAngelAGUIRRE](https://discourse.paraview.org/u/MiguelAngelAGUIRRE)
#### Post date: [March 19, 2020, 11:03am UTC](https://discourse.paraview.org/t/programmable-source-vorticity-and-gradient-from-numpy-data/3846/3 "2020-03-19T11:03:10Z")
Thanks for the info Mathieu !
I was wrapping the wrong vtk object. Here is the good version of the code for those that may have the same problem (Programmable source - vtk PolyData output).
Best regards
Miguel
```
# MODULES LOAD:
import numpy as np
import vtk
from vtk.numpy_interface import algorithms as algs
from vtk.numpy_interface import dataset_adapter as dsa
# NUMPY INPUT DATA:
X=np.array([0,1,2,0,1,2,0,1,2]) # X-coordinate
Y=np.array([0,0,0,1,1,1,2,2,2])
Z=np.array([0,0,0,0,0,0,0,0,0])
Vx=np.array([0,1,0,0,0,0,0,-1,0]) #X-velocity
Vy=np.array([0,0,0,-1,0,1,0,0,0])
Vz=np.array([0,0,0,0,0,0,0,0,0])
# DELAUNAY PLANE CREATION:
points = vtk.vtkPoints()
for i in range(X.size):
points.InsertNextPoint(X[i],Y[i],Z[i])
aPolyData = vtk.vtkPolyData()
aPolyData.SetPoints(points)
aCellArray = vtk.vtkCellArray()
boundary = vtk.vtkPolyData()
boundary.SetPoints(aPolyData.GetPoints())
boundary.SetPolys(aCellArray)
delaunay = vtk.vtkDelaunay2D()
if vtk.VTK_MAJOR_VERSION <= 5:
delaunay.SetInput(aPolyData.GetOutput())
delaunay.SetSource(boundary)
else:
delaunay.SetInputData(aPolyData)
delaunay.SetSourceData(boundary)
delaunay.Update()
self.GetOutput().ShallowCopy(delaunay.GetOutput())
# WRAP THE VTK ELEMENT:
WrapData = dsa.WrapDataObject(delaunay.GetOutput())
WrapData.PointData.append(Vx, "Vx")
WrapData.PointData.append(Vy, "Vy")
WrapData.PointData.append(Vz, "Vz")
Vx = WrapData.PointData["Vx"]
Vy = WrapData.PointData["Vy"]
Vz = WrapData.PointData["Vz"]
# GRADIENT AND VORTICITY:
Vector_Speed = algs.make_vector(Vx,Vy,Vz)
VORTICITY = algs.vorticity(Vector_Speed)
Vx_GRADIENT = algs.gradient(Vx)
# OUTPUT DATA:
output.PointData.append(Vector_Speed, "Vector_Speed")
output.PointData.append(VORTICITY, "VORTICITY")
output.PointData.append(Vx_GRADIENT, "Vx_GRADIENT")
```