The aim of blendvis is to us simple Python syntax for scientific plotting
with the full 3D rendering (and animation) capabilities of Blender.
The library uses similar commands to the matplotlib library and uses standard numerical data types such as numpy arrays.
The data is converted to an appropriate Blender scene and rendered using ray-tracing for high-quality 3D images.
We start with some simple examples of blendvis plots.
import blendvis
# simulate some experimental and theoretical data
M, N, H = 10, 10, 30
x, y = np.arange(1, M), np.arange(1, N)
X, Y = np.meshgrid(x, y)
Zth = H * np.exp( - (X * np.flip(Y))**2 / 10 )
Zex = Zth * np.random.uniform(0.8, 1, Zth.shape)
## create a figure axes
ax = blendvis.axes.Axes(verbose=False, use_latex=False)
# plot the data
ax.bar(X, Y, Zex, mat=mat, **dict(bevel=True, wireframe=False))
ax.bar(X, Y, Zth, mat='black', **dict(bevel=False, wireframe=True))
# change the look of the plot
ax.set(xlim=[0, M], ylim=[0, N], zlim=[0, H])
ax.set_xticklabels(xticks=x)
ax.set_yticklabels(yticks=y)
ax.set_zticklabels(zticks=np.linspace(0, H, 3).astype('int'))
ax.grid = False
ax.show()
import blendvis
# create figure axes
ax = blendvis.axes.Axes(verbose=False)
# generate data
num, points = 10, 100
theta = np.linspace(0, 2*np.pi, points)
r = np.linspace(0, 5, points)
# plot each line
for step in range(1, num, 1):
x = np.linspace(0, num, 20)
y = np.ones_like(x) * step
z = np.cumsum(np.abs(np.random.normal(0.1, 0.2, x.shape)))
ax.line(x=x, y=y, z=z, mat=flat)
# change the look
ax.set(xlim=[0, num], ylim=[0, num], zlim=[0, 5])
ax.set_xticklabels(xticks=np.arange(0, num, 1))
ax.set_yticklabels(yticks=np.arange(0, num, 1))
ax.set_zticklabels(zticks=np.linspace(0, 5, 6))
ax.show()
import blendvis
# create figure axes
ax = blendvis.axes.Axes(verbose=False)
# simulate data
num, range = 40, 10
x = np.random.uniform(0, range, num)
y = np.random.uniform(0, range, num)
z = np.random.uniform(0, range, num)
s = np.random.uniform(0, 0.5, num)
# plot data
ax.scatter(x, y, z, s, mat=mat)
# change look
ax.set(xlim=[0, range], ylim=[0, range], zlim=[0, range])
ax.set_xticklabels(xticks=np.arange(0, range, 1))
ax.set_yticklabels(yticks=np.arange(0, range, 1))
ax.set_zticklabels(zticks=np.linspace(0, range, 6))
ax.show()
blendvis must be run via an instance of Blender.
First download and install Blender (tested on 2.93 LTS version).
Download the source code and install as an add-on in Blender (Edit -> Preferences -> Add-ons -> Install).
Currently blendvis enables:
- bar, scatter, and line plots
- various colors, materials, and color maps
- LaTeX rendering (further testing needed)
- choose any system font
In the future, I hope to add:
- surface plots, tri plots
- subplots