# convert VTK array to numpy array
**URL:** https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152
**Category:** Support
**Created:** [April 25, 2020, 7:30am UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152 "2020-04-25T07:30:21Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@Deepa](https://discourse.vtk.org/u/Deepa)
#### Post date: [April 25, 2020, 7:30am UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152/1 "2020-04-25T07:30:21Z")
I have a VTKArray
`VTKArray([-3.17113447, -16.9386692 , 16.73578644], dtype=float32)`
I’m trying to convert this to a numpy array
`from vtk.util.numpy_support import vtk_to_numpy`
`vtk_to_numpy(edgeNode0_pos)`
error:
(self. **class**. **name** , name))
AttributeError: ‘VTKArray’ object has no attribute ‘GetDataType’
Any help on how to resolve this error will be useful.
---
### Author:  [@banesullivan](https://discourse.vtk.org/u/banesullivan)
#### Post date: [April 27, 2020, 11:45am UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152/2 "2020-04-27T11:45:15Z")
That `VTKArray` is a numpy array - its a helper class that wraps the `numpy.ndarray` class. `vtk.util.numpy_support.vtk_to_numpy` converts `vtkDataArray`s into `VTKArrays`.
Try:
```python
isinstance(edgeNode0_pos, np.ndarray)
```
---
### Author:  [@arfe](https://discourse.vtk.org/u/arfe)
#### Post date: [June 2, 2021, 2:58pm UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152/3 "2021-06-02T14:58:50Z")
I would like to convert the VTKArray into a numpy array anyway. Like a proper numpy array, no wrappers. Is that possible? I don’t understand how isinstance helps here.
Sorry for piggybacking on this old question, it just exactly describes what I’m looking for.
---
### Author:  [@zhang-qiang-github](https://discourse.vtk.org/u/zhang-qiang-github)
#### Post date: [June 2, 2021, 4:14pm UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152/4 "2021-06-02T16:14:47Z")
The following two functions may help.
```auto
def vtkToNumpy(data):
temp = vtk_to_numpy(data.GetPointData().GetScalars())
dims = data.GetDimensions()
component = data.GetNumberOfScalarComponents()
if component == 1:
numpy_data = temp.reshape(dims[2], dims[1], dims[0])
numpy_data = numpy_data.transpose(2,1,0)
elif component == 3 or component == 4:
if dims[2] == 1: # a 2D RGB image
numpy_data = temp.reshape(dims[1], dims[0], component)
numpy_data = numpy_data.transpose(0, 1, 2)
numpy_data = np.flipud(numpy_data)
else:
raise RuntimeError('unknow type')
return numpy_data
def numpyToVTK(data, multi_component=False, type='float'):
'''
multi_components: rgb has 3 components
type：float or char
'''
if type == 'float':
data_type = vtk.VTK_FLOAT
elif type == 'char':
data_type = vtk.VTK_UNSIGNED_CHAR
else:
raise RuntimeError('unknown type')
if multi_component == False:
if len(data.shape) == 2:
data = data[:, :, np.newaxis]
flat_data_array = data.transpose(2,1,0).flatten()
vtk_data = numpy_to_vtk(num_array=flat_data_array, deep=True, array_type=data_type)
shape = data.shape
else:
assert len(data.shape) == 3, 'only test for 2D RGB'
flat_data_array = data.transpose(1, 0, 2)
flat_data_array = np.reshape(flat_data_array, newshape=[-1, data.shape[2]])
vtk_data = numpy_to_vtk(num_array=flat_data_array, deep=True, array_type=data_type)
shape = [data.shape[0], data.shape[1], 1]
img = vtk.vtkImageData()
img.GetPointData().SetScalars(vtk_data)
img.SetDimensions(shape[0], shape[1], shape[2])
return img
```
---
### Author:  [@okahilak](https://discourse.vtk.org/u/okahilak)
#### Post date: [January 12, 2022, 1:14pm UTC](https://discourse.vtk.org/t/convert-vtk-array-to-numpy-array/3152/5 "2022-01-12T13:14:55Z")
Converting the VTKArray into a proper numpy array can done as follows:
```auto
>>> points
VTKArray([[7.881, 99.893, 31.901]], dtype=float32)
>>> import numpy as np
>>> np.array(points)
array([[7.881, 99.893, 31.901]], dtype=float32)
```