Python:根据点值制作一致的 3D 图
Python: make consistent 3D plot out of point values
在 2D 平面中,我有一组 nxn
点,由它们的 (x,y)
坐标定义。对于每个点,都存在一定数量,我想将其绘制为 3D 表面。我怎样才能在图中为每个点分配相关值?我问是因为我的情节很乱。
我会更好地解释自己。我有:
- 平面中
10x10=100
个点的位置字典:
dict1={0:(0, 0), 1:(0, 1), 2:(0, 2), ..., 99:(9,9)}
- 将与所述点关联的值字典:
dict2=OrderedDict([(0, 369670), (1, 370622), (2, 267034), ..., (99, 217500)])
dict1
和 dict2
的合并,其中每个值都与正确的点相关联,每个点都根据其坐标进行标记:
merged_dict={dict1[k]: v for k, v in dict2.items()}
merged_dict={(0,0):369670, (0,1):370622, (0,2):267034, ..., (9,9): 217500}
预期的 3D 图具有来自 merged_dict
的 X、Y 和 Z 点坐标。这是我的尝试:
#3D plot
import numpy as np
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
fig = plt.figure()
ax = fig.gca(projection='3d')
inds=[(0, 0), (0, 1), (0, 2), ..., (9,9)] #The coordinates of each point -> len(inds)=100
X=[]
for k in range(len(inds)):
X.append(int(inds[k][0]))
Y=X
X, Y = np.meshgrid(X, Y)
merged_dict = {dict1[k]: v for k, v in dict2.items()}
Z = merged_dict.values()
surf = ax.plot_surface(X, Y, Z, rstride=1, cstride=1, cmap=cm.winter, linewidth=0, antialiased=True)
ax.set_zlim(0, 900000)
ax.zaxis.set_major_locator(LinearLocator(10))
ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
plt.show()
我得到的图是错误的,因为它的表面很乱。预期结果是一个对称的钟形表面,峰值超过 800,000,与点 (4,4)、(5,4)、(4,5)、(5,5) 相对应。相反,结果让我认为 Z 值与相关的 X、Y 坐标没有正确关联。 如何解决这个问题?
编辑
这些是实际涉及的数据:
In[1]: merged_dict
Out[1]:
{(0, 0): 369670,
(0, 1): 370622,
(0, 2): 267034,
(0, 3): 169500,
(0, 4): 116014,
(0, 5): 116014,
(0, 6): 169500,
(0, 7): 267034,
(0, 8): 370622,
(0, 9): 369670,
(1, 0): 370622,
(1, 1): 491950,
(1, 2): 456750,
(1, 3): 370180,
(1, 4): 308118,
(1, 5): 308118,
(1, 6): 370180,
(1, 7): 456750,
(1, 8): 491950,
(1, 9): 370622,
(2, 0): 267034,
(2, 1): 456750,
(2, 2): 542718,
(2, 3): 554980,
(2, 4): 543588,
(2, 5): 543588,
(2, 6): 554980,
(2, 7): 542718,
(2, 8): 456750,
(2, 9): 267034,
(3, 0): 169500,
(3, 1): 370180,
(3, 2): 554980,
(3, 3): 689848,
(3, 4): 759272,
(3, 5): 759272,
(3, 6): 689848,
(3, 7): 554980,
(3, 8): 370180,
(3, 9): 169500,
(4, 0): 116014,
(4, 1): 308118,
(4, 2): 543588,
(4, 3): 759272,
(4, 4): 888268,
(4, 5): 888268,
(4, 6): 759272,
(4, 7): 543588,
(4, 8): 308118,
(4, 9): 116014,
(5, 0): 116014,
(5, 1): 308118,
(5, 2): 543588,
(5, 3): 759272,
(5, 4): 888268,
(5, 5): 888268,
(5, 6): 759272,
(5, 7): 543588,
(5, 8): 308118,
(5, 9): 116014,
(6, 0): 169500,
(6, 1): 370180,
(6, 2): 554980,
(6, 3): 689848,
(6, 4): 759272,
(6, 5): 759272,
(6, 6): 689848,
(6, 7): 554980,
(6, 8): 370180,
(6, 9): 169500,
(7, 0): 267034,
(7, 1): 456750,
(7, 2): 542718,
(7, 3): 554980,
(7, 4): 543588,
(7, 5): 543588,
(7, 6): 554980,
(7, 7): 542718,
(7, 8): 456750,
(7, 9): 267034,
(8, 0): 370622,
(8, 1): 491950,
(8, 2): 456750,
(8, 3): 370180,
(8, 4): 308118,
(8, 5): 308118,
(8, 6): 370180,
(8, 7): 456750,
(8, 8): 491950,
(8, 9): 370622,
(9, 0): 369670,
(9, 1): 370622,
(9, 2): 267034,
(9, 3): 169500,
(9, 4): 116014,
(9, 5): 116014,
(9, 6): 169500,
(9, 7): 267034,
(9, 8): 370622,
(9, 9): 369670}
我猜 Y=X
行有问题。 X, Y
是列表。当你写 Y=X
时,它意味着 Y
是对 X
的引用。您需要复制一份,即:
Y = []
numpy.copy(Y,X)
测试这个变体并写下会发生什么。否则 post merged_dict 的数据用于测试实际值的图。
更新:
看图,对吗?
数组 Z
必须是二维数组,但在您的代码中它是一维列表。
import matplotlib.pyplot as plt
import numpy as np
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
fig = plt.figure()
ax = fig.gca(projection='3d')
X = np.arange(0,10,1)
Y = np.arange(0,10,1)
X, Y = np.meshgrid(X, Y)
merged_dict = {(0, 0): 369670,
(0, 1): 370622,
(0, 2): 267034,
...
(9, 8): 370622,
(9, 9): 369670}
Z = np.array(merged_dict.values()).reshape(10,10)
surf = ax.plot_surface(X, Y, Z, rstride=1, cstride=1, cmap=cm.winter, linewidth=0, antialiased=True)
ax.set_zlim(0, 900000)
plt.show()
更新 2:
问题是因为你的数据存储在字典中,但字典没有按索引排序。将值加载到数组 Z
:
Z = np.zeros((10,10))
for key in merged_dict:
i = key[0]
j = key[1]
Z[i][j] = int(merged_dict[key])
因此,您得到:
在 2D 平面中,我有一组 nxn
点,由它们的 (x,y)
坐标定义。对于每个点,都存在一定数量,我想将其绘制为 3D 表面。我怎样才能在图中为每个点分配相关值?我问是因为我的情节很乱。
我会更好地解释自己。我有:
- 平面中
10x10=100
个点的位置字典:
dict1={0:(0, 0), 1:(0, 1), 2:(0, 2), ..., 99:(9,9)}
- 将与所述点关联的值字典:
dict2=OrderedDict([(0, 369670), (1, 370622), (2, 267034), ..., (99, 217500)])
dict1
和dict2
的合并,其中每个值都与正确的点相关联,每个点都根据其坐标进行标记:
merged_dict={dict1[k]: v for k, v in dict2.items()}
merged_dict={(0,0):369670, (0,1):370622, (0,2):267034, ..., (9,9): 217500}
预期的 3D 图具有来自 merged_dict
的 X、Y 和 Z 点坐标。这是我的尝试:
#3D plot
import numpy as np
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
fig = plt.figure()
ax = fig.gca(projection='3d')
inds=[(0, 0), (0, 1), (0, 2), ..., (9,9)] #The coordinates of each point -> len(inds)=100
X=[]
for k in range(len(inds)):
X.append(int(inds[k][0]))
Y=X
X, Y = np.meshgrid(X, Y)
merged_dict = {dict1[k]: v for k, v in dict2.items()}
Z = merged_dict.values()
surf = ax.plot_surface(X, Y, Z, rstride=1, cstride=1, cmap=cm.winter, linewidth=0, antialiased=True)
ax.set_zlim(0, 900000)
ax.zaxis.set_major_locator(LinearLocator(10))
ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
plt.show()
我得到的图是错误的,因为它的表面很乱。预期结果是一个对称的钟形表面,峰值超过 800,000,与点 (4,4)、(5,4)、(4,5)、(5,5) 相对应。相反,结果让我认为 Z 值与相关的 X、Y 坐标没有正确关联。 如何解决这个问题?
编辑
这些是实际涉及的数据:
In[1]: merged_dict
Out[1]:
{(0, 0): 369670,
(0, 1): 370622,
(0, 2): 267034,
(0, 3): 169500,
(0, 4): 116014,
(0, 5): 116014,
(0, 6): 169500,
(0, 7): 267034,
(0, 8): 370622,
(0, 9): 369670,
(1, 0): 370622,
(1, 1): 491950,
(1, 2): 456750,
(1, 3): 370180,
(1, 4): 308118,
(1, 5): 308118,
(1, 6): 370180,
(1, 7): 456750,
(1, 8): 491950,
(1, 9): 370622,
(2, 0): 267034,
(2, 1): 456750,
(2, 2): 542718,
(2, 3): 554980,
(2, 4): 543588,
(2, 5): 543588,
(2, 6): 554980,
(2, 7): 542718,
(2, 8): 456750,
(2, 9): 267034,
(3, 0): 169500,
(3, 1): 370180,
(3, 2): 554980,
(3, 3): 689848,
(3, 4): 759272,
(3, 5): 759272,
(3, 6): 689848,
(3, 7): 554980,
(3, 8): 370180,
(3, 9): 169500,
(4, 0): 116014,
(4, 1): 308118,
(4, 2): 543588,
(4, 3): 759272,
(4, 4): 888268,
(4, 5): 888268,
(4, 6): 759272,
(4, 7): 543588,
(4, 8): 308118,
(4, 9): 116014,
(5, 0): 116014,
(5, 1): 308118,
(5, 2): 543588,
(5, 3): 759272,
(5, 4): 888268,
(5, 5): 888268,
(5, 6): 759272,
(5, 7): 543588,
(5, 8): 308118,
(5, 9): 116014,
(6, 0): 169500,
(6, 1): 370180,
(6, 2): 554980,
(6, 3): 689848,
(6, 4): 759272,
(6, 5): 759272,
(6, 6): 689848,
(6, 7): 554980,
(6, 8): 370180,
(6, 9): 169500,
(7, 0): 267034,
(7, 1): 456750,
(7, 2): 542718,
(7, 3): 554980,
(7, 4): 543588,
(7, 5): 543588,
(7, 6): 554980,
(7, 7): 542718,
(7, 8): 456750,
(7, 9): 267034,
(8, 0): 370622,
(8, 1): 491950,
(8, 2): 456750,
(8, 3): 370180,
(8, 4): 308118,
(8, 5): 308118,
(8, 6): 370180,
(8, 7): 456750,
(8, 8): 491950,
(8, 9): 370622,
(9, 0): 369670,
(9, 1): 370622,
(9, 2): 267034,
(9, 3): 169500,
(9, 4): 116014,
(9, 5): 116014,
(9, 6): 169500,
(9, 7): 267034,
(9, 8): 370622,
(9, 9): 369670}
我猜 Y=X
行有问题。 X, Y
是列表。当你写 Y=X
时,它意味着 Y
是对 X
的引用。您需要复制一份,即:
Y = []
numpy.copy(Y,X)
测试这个变体并写下会发生什么。否则 post merged_dict 的数据用于测试实际值的图。
更新:
看图,对吗?
数组 Z
必须是二维数组,但在您的代码中它是一维列表。
import matplotlib.pyplot as plt
import numpy as np
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
fig = plt.figure()
ax = fig.gca(projection='3d')
X = np.arange(0,10,1)
Y = np.arange(0,10,1)
X, Y = np.meshgrid(X, Y)
merged_dict = {(0, 0): 369670,
(0, 1): 370622,
(0, 2): 267034,
...
(9, 8): 370622,
(9, 9): 369670}
Z = np.array(merged_dict.values()).reshape(10,10)
surf = ax.plot_surface(X, Y, Z, rstride=1, cstride=1, cmap=cm.winter, linewidth=0, antialiased=True)
ax.set_zlim(0, 900000)
plt.show()
更新 2:
问题是因为你的数据存储在字典中,但字典没有按索引排序。将值加载到数组 Z
:
Z = np.zeros((10,10))
for key in merged_dict:
i = key[0]
j = key[1]
Z[i][j] = int(merged_dict[key])
因此,您得到: