3D curves in a single plot about pyvista-support

[GuillaumeFavelier]

With this totally unoptimized snippet:

import numpy as np
import pyvista as pv
import csv
from pyvista.utilities import lines_from_points

xfile = open('x_coordinates.csv', 'r')
xreader = csv.reader(xfile, delimiter=',')
x = []
for row in xreader:
    x.append(row)
xfile.close()

yfile = open('y_coordinates.csv', 'r')
yreader = csv.reader(yfile, delimiter=',')
y = []
for row in yreader:
    y.append(row)
yfile.close()

zfile = open('z_coordinates.csv', 'r')
zreader = csv.reader(zfile, delimiter=',')
z = []
for row in zreader:
    z.append(row)
zfile.close()

plotter = pv.Plotter()
plotter.background_color = 'white'
for idx in range(len(x)):
    cx = x[idx]
    cy = y[idx]
    cz = z[idx]
    pts = np.column_stack((cx, cy, cz)).astype(np.float)
    line = lines_from_points(pts)
    plotter.add_mesh(line, show_edges=True)
plotter.show()

I got the following:

Did you expect something similar?

Also, I think it should be possible to append/create an UnstructuredGrid dataset and use only one call to add_mesh(). Note that I had to use show_edges=True to actually visualize the result.

[Sunil7545]

Dear Guillaume, Thank you very much for your quick response. It should look like this. Is it possible to add colours to these curves (Z height as a scalar field) similar to streamlines and make it tube-like structure (like streamlines)? Regards Sunil

…

On Tue, Jun 4, 2019 at 2:52 PM Guillaume Favelier ***@***.***> wrote: With this totally unoptimized snippet: import numpy as npimport pyvista as pvimport csvfrom pyvista.utilities import lines_from_points xfile = open('x_coordinates.csv', 'r') xreader = csv.reader(xfile, delimiter=',') x = []for row in xreader: x.append(row) xfile.close() yfile = open('y_coordinates.csv', 'r') yreader = csv.reader(yfile, delimiter=',') y = []for row in yreader: y.append(row) yfile.close() zfile = open('z_coordinates.csv', 'r') zreader = csv.reader(zfile, delimiter=',') z = []for row in zreader: z.append(row) zfile.close() plotter = pv.Plotter() plotter.background_color = 'white'for idx in range(len(x)): cx = x[idx] cy = y[idx] cz = z[idx] pts = np.column_stack((cx, cy, cz)).astype(np.float) line = lines_from_points(pts) plotter.add_mesh(line, show_edges=True) plotter.show() I got the following: [image: image] <https://user-images.githubusercontent.com/18143289/58880247-cc963f00-86d7-11e9-8da5-7fa2cb0fa0bb.png> Did you expect something similar? Also, I think it should be possible to append/create an UnstructuredGrid dataset and use only one call to add_mesh(). Note that I had to use show_edges=True to actually visualize the result. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#11?email_source=notifications&email_token=AEB5O7UOZVWS7UBLJLK2SVLPYZQRHA5CNFSM4HS2PI5KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW4OZSI#issuecomment-498658505>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEB5O7QLXMX2UUQD5WH6ZQ3PYZQRHANCNFSM4HS2PI5A> .

-- Sunil Kumar Yadav Freie University Berlin Berlin, Germany Email ID: *[email protected] <[email protected]>* Tel(Office) : [phone] Mob: [phone]

[GuillaumeFavelier]

I modified a bit the previous (unoptimized) snippet:

...
plotter = pv.Plotter()
plotter.background_color = 'white'
for idx in range(len(x)):
    cx = x[idx]
    cy = y[idx]
    cz = z[idx]
    pts = np.column_stack((cx, cy, cz)).astype(np.float)

    lines = lines_from_points(pts)
    lines.lines = lines.faces
    tube = lines.tube(0.01).elevation()
    plotter.add_mesh(tube, show_edges=False)
plotter.show()

The main modifications are lines.lines = lines.faces which basically updates the lines of the PolyData required by the tube filter. The rest is pretty straightforward, we apply the filters: first the tube with a radius of 0.01 then the elevation filter to add a scalar field corresponding to the Z height.

There is maybe a better/efficient way to achieve this, what do you think @pyvista/developers ?

[Sunil7545]

Dear Guillaume, Thank you very much. I was looking for an exactly similar structure. I will try to optimize it. Regards Sunil

…

On Tue, Jun 4, 2019 at 5:30 PM Guillaume Favelier ***@***.***> wrote: I modified a bit the previous (unoptimized) snippet: ... plotter = pv.Plotter() plotter.background_color = 'white'for idx in range(len(x)): cx = x[idx] cy = y[idx] cz = z[idx] pts = np.column_stack((cx, cy, cz)).astype(np.float) lines = lines_from_points(pts) lines.lines = lines.faces tube = lines.tube(0.01).elevation() plotter.add_mesh(tube, show_edges=False) plotter.show() [image: image] <https://user-images.githubusercontent.com/18143289/58891797-7633fb00-86ed-11e9-9d28-ccbed2797000.png> The main modifications are lines.lines = lines.faces which basically updates the lines of the PolyData required by the tube filter. The rest is pretty straightforward, we apply the filters: first the tube with a radius of 0.01 then the elevation filter to add a scalar field corresponding to the Z height. There is maybe a better/efficient way to achieve this, what do you think @pyvista/developers <https://github.com/orgs/pyvista/teams/developers> ? — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#11?email_source=notifications&email_token=AEB5O7Q24BMTBLQEXORMAOLPY2DADA5CNFSM4HS2PI5KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW46LGI#issuecomment-498722201>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEB5O7TV32AVPEPRVKMMSZTPY2DADANCNFSM4HS2PI5A> .

-- Sunil Kumar Yadav Freie University Berlin Berlin, Germany Email ID: *[email protected] <[email protected]>* Tel(Office) : [phone] Mob: [phone]

[akaszynski]

Good start @GuillaumeFavelier!

Here's a little bit of an improvement using splines, which leads to smoother plotting with no breaks along the line.

I'm adding splines to pyvista for the next release, so you won't need to use make_spline and will instead use pv.Spline

import numpy as np
import pyvista as pv

x = np.loadtxt('x_coordinates.csv', delimiter=',')
y = np.loadtxt('y_coordinates.csv', delimiter=',')
z = np.loadtxt('z_coordinates.csv', delimiter=',')


def make_spline(points, n_points=None):
    """Create a spline from points

    Parameters
    ----------
    points : np.ndarray
        Array of points to build a spline out of.  Array must be 3D
        and directionally ordered.

    n_points : int, optional
        Number of points to interpolate along the points array.

    Returns
    -------
    spline : pyvista.PolyData
        Line mesh of spline.

    Examples
    --------
    Construct a spline
    >>> import numpy as np
    >>> import pyvista as pv
    >>> theta = np.linspace(-4 * np.pi, 4 * np.pi, 100)
    >>> z = np.linspace(-2, 2, 100)
    >>> r = z**2 + 1
    >>> x = r * np.sin(theta)
    >>> y = r * np.cos(theta)
    >>> points = np.column_stack((x, y, z))
    >>> spline = pv.Spline(points, 1000)
    """
    import vtk
    spline = vtk.vtkParametricSpline()
    vtk_points = pv.vtk_points(points, False)
    spline.SetPoints(vtk_points)
    spline.GetPoints().Modified()
    spline.Modified()

    # get interpolation density
    u_res = n_points
    if u_res is None:
        u_res = points.shape[0]

    return pv.geometric_objects.surface_from_para(spline, u_res)


# convert points to splines
splines = []
for idx in range(x.shape[0]):
    points = np.column_stack((x[idx], y[idx], z[idx]))

    spline = make_spline(points)
    # spline = pv.Spline(points)  # in the next release
    splines.append(spline)


# boring plot
pv.plot(splines)

# exciting plot with scalars
tubes = []
for spline in splines:
    # add scalars here...
    spline['scalars'] = np.arange(spline.n_points)
    tubes.append(spline.tube(0.01))

pv.plot(tubes, smooth_shading=True)

[Sunil7545]

Dear Alex, Thank you very much for this elegant solution. The picture looks really good. However, I am not able to run this code. I am getting the following error: return pv.geometric_objects.surface_from_para(spline, u_res) AttributeError: module 'pyvista.geometric_objects' has no attribute 'surface_from_para' Please take a look. Regards Sunil

…

On Wed, Jun 5, 2019 at 9:30 AM Alex Kaszynski ***@***.***> wrote: Good start @GuillaumeFavelier <https://github.com/GuillaumeFavelier>! Here's a little bit of an improvement using splines, which leads to smoother plotting with no breaks along the line. I'm adding splines to pyvista for the next release, so you won't need to use make_spline and will instead use pv.Spline import numpy as npimport pyvista as pv x = np.loadtxt('x_coordinates.csv', delimiter=',') y = np.loadtxt('y_coordinates.csv', delimiter=',') z = np.loadtxt('z_coordinates.csv', delimiter=',') def make_spline(points, n_points=None): """Create a spline from points Parameters ---------- points : np.ndarray Array of points to build a spline out of. Array must be 3D and directionally ordered. n_points : int, optional Number of points to interpolate along the points array. Returns ------- spline : pyvista.PolyData Line mesh of spline. Examples -------- Construct a spline >>> import numpy as np >>> import pyvista as pv >>> theta = np.linspace(-4 * np.pi, 4 * np.pi, 100) >>> z = np.linspace(-2, 2, 100) >>> r = z**2 + 1 >>> x = r * np.sin(theta) >>> y = r * np.cos(theta) >>> points = np.column_stack((x, y, z)) >>> spline = pv.Spline(points, 1000) """ import vtk spline = vtk.vtkParametricSpline() vtk_points = pv.vtk_points(points, False) spline.SetPoints(vtk_points) spline.GetPoints().Modified() spline.Modified() # get interpolation density u_res = n_points if u_res is None: u_res = points.shape[0] return pv.geometric_objects.surface_from_para(spline, u_res) # convert points to splines splines = []for idx in range(x.shape[0]): points = np.column_stack((x[idx], y[idx], z[idx])) spline = make_spline(points) # spline = pv.Spline(points) # in the next release splines.append(spline) # boring plot pv.plot(splines) # exciting plot with scalars tubes = []for spline in splines: # add scalars here... spline['scalars'] = np.arange(spline.n_points) tubes.append(spline.tube(0.01)) pv.plot(tubes, smooth_shading=True) [image: cool] <https://user-images.githubusercontent.com/11981631/58937943-1a608500-8774-11e9-8f26-c1da1ad0044f.png> — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#11?email_source=notifications&email_token=AEB5O7R3JO2L3OGIS3NC2ALPY5TSBA5CNFSM4HS2PI5KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW63FDA#issuecomment-498971276>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEB5O7XWHGYSBKC7RFV2MHDPY5TSBANCNFSM4HS2PI5A> .

-- Sunil Kumar Yadav Freie University Berlin Berlin, Germany Email ID: *[email protected] <[email protected]>* Tel(Office) : [phone] Mob: [phone]

[akaszynski]

Are you running the latest pyvista?

[Sunil7545]

Thank you, I wasn't using the latest version. Everything is perfect now.

…

On Wed, Jun 5, 2019 at 10:16 AM Alex Kaszynski ***@***.***> wrote: Are you running the latest pyvista? — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#11?email_source=notifications&email_token=AEB5O7WICO5WYM4WAICFVXLPY5Y6HA5CNFSM4HS2PI5KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW66SCA#issuecomment-498985224>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEB5O7UG2YGRBXHMHQWCZYTPY5Y6HANCNFSM4HS2PI5A> .

-- Sunil Kumar Yadav Freie University Berlin Berlin, Germany Email ID: *[email protected] <[email protected]>* Tel(Office) : [phone] Mob: [phone]

[Sunil7545]

Dear Alex, Can I add a mesh geometry to the same panel ? e.g. using add_mesh?

…

On Wed, Jun 5, 2019 at 10:16 AM Alex Kaszynski ***@***.***> wrote: Are you running the latest pyvista? — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#11?email_source=notifications&email_token=AEB5O7WICO5WYM4WAICFVXLPY5Y6HA5CNFSM4HS2PI5KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW66SCA#issuecomment-498985224>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEB5O7UG2YGRBXHMHQWCZYTPY5Y6HANCNFSM4HS2PI5A> .

-- Sunil Kumar Yadav Freie University Berlin Berlin, Germany Email ID: *[email protected] <[email protected]>* Tel(Office) : [phone] Mob: [phone]

[akaszynski]

Not sure what you’re trying to do. Can you provide a code example?