极端基础知识：将数组显示为图像（2d或3d）

Question

I'm completely new to Numpy, and I have not used Python in almost 8 years [so it's best to assume I'm completely new to Python, too].

I am trying to display a 2-dimensional array as a colorful image. I would like to be able to do this with 3-dimensional arrays, too.

For context, I would like to display this array ( array with letters ) with colors like this: ( array with colors )

Step two would be to be able to make a rotatable figure that would show a 3d array (essentially like array 1 above but with an extra ABCD dimension, making triples like ABC, ABD, ACD, etc. instead of pairs like AB, AC, AD etc.).

Answer 1

Important to note: I am not completely certain whether I understood your request right; hence I'll give a answer first, which covers your question only partly. I'll improve this later, once you let me know I'm going in the right direction.

#!/usr/bin/env python3

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.colors import ListedColormap

### create a dictionary with:
#   dictionary keys: will be used as color index
#   "name": "AB" <-- the text value
#   "color": "#ff0000" <-- a hex color value to represent each square
#   "coordinate": [1,0] <-- coordinates as x in (0,1,2) 
#                           and y-index (0,1) : where to plot this entry
dataDict={
    0:{"name":"AA",
    "color":"#ff0000",
    "coordinate":[0,0]},
    1:{"name":"AB",
    "color":"#00ff00",
    "coordinate":[1,0]},
    2:{"name":"AC",
    "color":"#0000ff",
    "coordinate":[2,0]},
    3:{"name":"BA",
    "color":"#0fff0f",
    "coordinate":[0,1]},
    4:{"name":"BB",
    "color":"#cecece",
    "coordinate":[1,1]},
    5:{"name":"BC",
    "color":"#000033",
    "coordinate":[2,1]},
}

### define the size of your array in x- and y-direction
x_size=3
y_size=2

### create an empty image array of proper dimensions
img_array = np.zeros((y_size,x_size))

### iterate over the dictionary: 
#   - looking up the color index (0-5)
#   - and store it in the img_array
for i,v in dataDict.items():
    [xi,yi]=v["coordinate"]
    img_array[yi,xi] = i

### create a colormap which 
#   maps the dictionary keys (0-5) to the respective color value
cmap = ListedColormap([v["color"] for i,v in dataDict.items()])

### create a figure and subplot
fig,ax=plt.subplots(1,1)
### tell the subplot to show the image "img_array" using the colormap "cmap"
ax.imshow(img_array,cmap=cmap,zorder=1,origin="upper")

#### iterate over the dictionary, get the coordiantes and names, and place text
for i,v in dataDict.items():
    print(i,v["coordinate"][0],v["coordinate"][1])
    ax.text(v["coordinate"][0],v["coordinate"][1],v["name"],zorder=2,)

### shwo the plot    
plt.show()

Answer 2

In response to Asmus: This is a really good start. I tried to take your code and modify it, but unforunately something went wrong. Some of the cells are not the colors I'm telling them to be.

I'm also curious if it's possible to make a 3-dimensional version of this. This would have 3-d cells of different colors, each cell representing a different 3-letter permutation, where two cells have the same color if they contain the exact same 3 letters.

#!/usr/bin/env python3

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.colors import ListedColormap

### create a dictionary with:
#   dictionary keys: will be used as color index
#   "name": "AB" <-- the text value
#   "color": "#ff0000" <-- a hex color value to represent each square
#   "coordinate": [1,0] <-- coordinates as x in (0,1,2) 
#                           and y-index (0,1) : where to plot this entry

white = "#ffffff"
grey = "#cecece"
red = "#ff0000"
green = "#00ff00"
purple = "#ccbbee"
pink = "#ffaabb"
dataDict={
    1:{"name":"A",
    "color":white,
    "coordinate": [0, 1]},
    2:{"name":"B",
    "color":white,
    "coordinate": [0, 2]},
    3:{"name":"C",
    "color":white,
    "coordinate": [0, 3]},
    4:{"name":"A",
    "color":white,
    "coordinate": [1, 0]},
    5:{"name":"B",
    "color":white,
    "coordinate": [2, 0]},
    6:{"name":"C",
    "color":white,
    "coordinate": [3, 0]},

    7:{"name":"AA",
    "color":grey,
    "coordinate":[1,1]},

    8:{"name":"AB",
    "color":green,
    "coordinate":[2,1]},

    9:{"name":"AC",
    "color":pink,
    "coordinate":[3,1]},

    10:{"name":"BA",
    "color":green,
    "coordinate":[1,2]},

    11:{"name":"BB",
    "color":grey,
    "coordinate":[2,2]},

    12:{"name":"BC",
    "color":purple,
    "coordinate":[3,2]},

    13:{"name":"CA",
    "color":pink,
    "coordinate":[1,3]},

    14:{"name":"CB",
    "color":purple,
    "coordinate":[2,3]},

    15:{"name":"CC",
    "color":grey,
    "coordinate":[3,3]}
}

### define the size of your array in x- and y-direction
x_size=4
y_size=4

### create an empty image array of proper dimensions
img_array = np.zeros((y_size,x_size))

### iterate over the dictionary: 
#   - looking up the color index (0-5)
#   - and store it in the img_array
for i,v in dataDict.items():
    [xi,yi]=v["coordinate"]
    img_array[yi,xi] = i

### create a colormap which 
#   maps the dictionary keys (0-5) to the respective color value
cmap = ListedColormap([v["color"] for i,v in dataDict.items()])

### create a figure and subplot
fig,ax=plt.subplots(1,1)
### tell the subplot to show the image "img_array" using the colormap "cmap"
ax.imshow(img_array,cmap=cmap,zorder=1,origin="upper")

#### iterate over the dictionary, get the coordiantes and names, and place text
for i,v in dataDict.items():
    print(i,v["coordinate"][0],v["coordinate"][1])
    ax.text(v["coordinate"][0],v["coordinate"][1],v["name"],zorder=2,)

### shwo the plot    
plt.show()