如何在Google Colaboratory中用Python可视化3D核？求Mayavi替代库
我尝试在Google Colaboratory中用Python可视化3D核数组，一开始想用Mayavi库，但安装它依赖的wxPython时失败了。执行的安装命令：
pip install attrdict3 pip install requests pip install wxpython
出现的错误：
Building wheels for collected packages: wxpython
error: subprocess-exited-with-error× python setup.py bdist_wheel did not run successfully.
│ exit code: 1
╰─> See above for output.note: This error originates from a subprocess, and is likely not a problem with pip.
Building wheel for wxpython (setup.py) ... error
ERROR: Failed building wheel for wxpython
Running setup.py clean for wxpython
Failed to build wxpython
Installing collected packages: wxpython
error: subprocess-exited-with-error× Running setup.py install for wxpython did not run successfully.
│ exit code: 1
╰─> See above for output.note: This error originates from a subprocess, and is likely not a problem with pip.
Running setup.py install for wxpython ... error
error: legacy-install-failure× Encountered error while trying to install package.
╰─> wxpythonnote: This is an issue with the package mentioned above, not pip.
hint: See above for output from the failure.
我原本的可视化代码：
from mayavi import mlab mlab.volume_slice(kernel, plane_orientation='x_axes', slice_index=3) mlab.volume_slice(kernel, plane_orientation='y_axes', slice_index=3) mlab.volume_slice(kernel, plane_orientation='z_axes', slice_index=3) mlab.show() mlab.contour3d(kernel) mlab.show()
待可视化的3D核数组：
import numpy as np kernel = np.array([[[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]], [[0.88369477, 0.9201748 , 0.5593715 ], [0.91746247, 0.9553365 , 0.5807462 ], [0.88369477, 0.9201748 , 0.5593715 ]], [[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]]])
求推荐适合这个场景的其他可视化库。
以下几个库无需依赖wxPython，且能在Google Colab中正常运行，适配你的3D核数组可视化需求：
1. Matplotlib（基础3D可视化）
Matplotlib是Python最常用的可视化库，自带3D模块，能实现切片和等值面可视化，无需额外复杂依赖。
实现代码：
import numpy as np import matplotlib.pyplot as plt from mpl_toolkits.mplot3d import Axes3D # 定义核数组 kernel = np.array([[[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]], [[0.88369477, 0.9201748 , 0.5593715 ], [0.91746247, 0.9553365 , 0.5807462 ], [0.88369477, 0.9201748 , 0.5593715 ]], [[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]]]) # 1. 切片可视化 fig = plt.figure(figsize=(15,5)) # X轴切片（索引1，数组为3x3x3，索引范围0-2） ax1 = fig.add_subplot(131, projection='3d') x_slice = kernel[1, :, :] X, Y = np.meshgrid(range(3), range(3)) ax1.plot_surface(X, Y, x_slice, cmap='viridis') ax1.set_title('X轴切片 (index=1)') # Y轴切片 ax2 = fig.add_subplot(132, projection='3d') y_slice = kernel[:, 1, :] X, Z = np.meshgrid(range(3), range(3)) ax2.plot_surface(X, y_slice, Z, cmap='viridis') ax2.set_title('Y轴切片 (index=1)') # Z轴切片 ax3 = fig.add_subplot(133, projection='3d') z_slice = kernel[:, :, 1] X, Y = np.meshgrid(range(3), range(3)) ax3.plot_surface(X, Y, z_slice, cmap='viridis') ax3.set_title('Z轴切片 (index=1)') plt.show() # 2. 等值面可视化 fig = plt.figure() ax = fig.add_subplot(111, projection='3d') # 生成网格坐标 x, y, z = np.mgrid[0:3, 0:3, 0:3] # 绘制等值面（阈值设为0.8、0.9、0.95） ax.contour3D(x, y, z, kernel, levels=[0.8, 0.9, 0.95], cmap='viridis') ax.set_title('3D等值面') plt.show()
2. Plotly（交互式3D可视化）
Plotly支持交互式操作（旋转、缩放、悬停查看数值），在Colab中渲染效果良好，适合探索性分析。
实现代码：
import numpy as np import plotly.graph_objects as go kernel = np.array([[[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]], [[0.88369477, 0.9201748 , 0.5593715 ], [0.91746247, 0.9553365 , 0.5807462 ], [0.88369477, 0.9201748 , 0.5593715 ]], [[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]]]) # 1. 切片可视化（X轴切片，其他轴同理） x_slice = kernel[1, :, :] X, Y = np.meshgrid(range(3), range(3)) fig = go.Figure(data=[go.Surface(z=x_slice, x=X, y=Y)]) fig.update_layout(title='X轴切片 (index=1)', autosize=False, width=600, height=500) fig.show() # 2. 3D等值面可视化 x, y, z = np.mgrid[0:3, 0:3, 0:3] fig = go.Figure(data=go.Isosurface( x=x.flatten(), y=y.flatten(), z=z.flatten(), value=kernel.flatten(), isomin=0.8, isomax=0.95, colorscale='Viridis', surface_count=3, # 显示3个等值面 )) fig.update_layout(title='3D核数组等值面') fig.show()
3. PyVista（专业3D科学可视化）
PyVista专注于3D网格和体数据可视化，API简洁，能轻松实现切片、等值面等操作，适合处理三维数值数组。
实现代码：
!pip install pyvista -q # Colab中需要先安装 import numpy as np import pyvista as pv kernel = np.array([[[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]], [[0.88369477, 0.9201748 , 0.5593715 ], [0.91746247, 0.9553365 , 0.5807462 ], [0.88369477, 0.9201748 , 0.5593715 ]], [[0.88121283, 0.91759044, 0.5578005 ], [0.91488576, 0.9526534 , 0.5791151 ], [0.88121283, 0.91759044, 0.5578005 ]]]) # 将numpy数组转为PyVista的体数据 grid = pv.UniformGrid() grid.dimensions = np.array(kernel.shape) + 1 grid.spacing = (1, 1, 1) grid.origin = (0, 0, 0) grid.cell_data['values'] = kernel.flatten(order='F') # 1. 多平面切片可视化 plotter = pv.Plotter() plotter.add_mesh(grid.slice('x', origin=(1,1,1)), cmap='viridis') plotter.add_mesh(grid.slice('y', origin=(1,1,1)), cmap='viridis') plotter.add_mesh(grid.slice('z', origin=(1,1,1)), cmap='viridis') plotter.show() # 2. 等值面可视化 plotter = pv.Plotter() plotter.add_mesh(grid.contour([0.8, 0.9, 0.95]), cmap='viridis') plotter.show()
内容的提问来源于stack exchange，提问作者Defne
