Add legend of whole numbers instead of gradient in matplotlib - python

I'd like to add a legend that is only whole numbers, i.e. 0,1,2,3...14, instead of the gradient color bar. Basically, I want the array values to have a unique color and label in the legend, so that you can clearly distinguish each value in the array.
fileloc=os.path.join(basepath, infile)
data=np.loadtxt(fileloc)
fig = plt.figure(figsize=(20,10))
plt.imshow(data)
plt.colorbar()

If I understand the question correctly, the data is given as an integer numpy array which results in an image with exactly N different colors.
To get a colormap with exactly N colors from the viridis colormap, use plt.cm.get_cmap('viridis', N). This will result in a colorbar with exactly N regions.
To get ticks nicely in the center of each region, divide the space into 2N+1 pieces, and then take all the odd positions. (So, if there are 5 colors, the colorbar will go from 0 to 4, which would get 11 marks of which mark 0 is skipped and marks 1, 3, 5, 7 and 9 are used). Next to each tick a label with the number can be placed.
import numpy as np
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(6, 4))
# create some random test data
data = np.random.normal(0, 0.05, size=(150, 150)).cumsum(axis=0).cumsum(axis=1)
data = data.astype(np.int) # convert to integers
data -= data.min() # let the numbers start at zero
num_colors = data.max() + 1
cmap = plt.cm.get_cmap('viridis', num_colors)
plt.imshow(data, cmap=cmap)
cbar = plt.colorbar(ticks=np.linspace(0, num_colors - 1, num_colors * 2 + 1)[1::2])
cbar.ax.set_yticklabels(range(num_colors))
plt.show()

You can use matplotlib.colors.ListedColormap as follows:
custom_cmap = colors.ListedColormap(['purple','blue','green','yellow']) #... and so on until you have 15 colours specified
Then just pass that as the cmap argument to imshow and colorbar:
plt.imshow(data, cmap=custom_cmap)
plt.colorbar(cmap=custom_cmap)

Related

draw a color grid based on points density using python matplotlib

The question is to read 10,000 coordinate points from a file and create a colored grid based on the density of each block on the grid. The range of x-axis is [-73.59, -73.55] and the y-axis is [45.49,45.530]. My code will plot a grid with many different colors, now I need a feature to only color the grid that has a specific density n, for example, The n = 100, only the grid with 100 points or higher will be colored to yellow, and other grids will be black.
I just added a link to my shapefile
https://drive.google.com/open?id=1H-8FhfonnPrYW9y7RQZDtiNLxVEiC6R8
import numpy as np
import matplotlib.pyplot as plt
import shapefile
grid_size = 0.002
x1 = np.arange(-73.59,-73.55,grid_size)
y1 = np.arange(45.49,45.530,grid_size)
shape = shapefile.Reader("Shape/crime_dt.shp",encoding='ISO-8859-1')
shapeRecords = shape.shapeRecords()
x_coordinates=[]
y_coordinates=[]
# read all points in .shp file, and store them in 2 lists.
for k in range(len(shapeRecords)):
x = float(shapeRecords[k].shape.__geo_interface__["coordinates"][0])
y = float(shapeRecords[k].shape.__geo_interface__["coordinates"][1])
x_coordinates.append(x)
y_coordinates.append(y)
plt.hist2d(x_coordinates,y_coordinates,bins=[x1,y1])
plt.show()
You can create a colormap with just two colors, and set vmin and vmax to be symmetrical around your desired pivot value.
Optionally you put the value of each bin inside the cells, while the pivot value decides the text color.
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
grid_size = 0.002
x1 = np.arange(-73.59, -73.55, grid_size)
y1 = np.arange(45.49, 45.530, grid_size)
# read coordinates from file and put them into two lists, similar to this
x_coordinates = np.random.uniform(x1.min(), x1.max(), size=40000)
y_coordinates = np.random.uniform(y1.min(), y1.max(), size=40000)
pivot_value = 100
# create a colormap with two colors, vmin and vmax are chosen so that their center is the pivot value
cmap = ListedColormap(['indigo', 'gold'])
# create a 2d histogram with xs and ys as bin boundaries
binvalues, _, _, _ = plt.hist2d(x_coordinates, y_coordinates, bins=[x1, y1], cmap=cmap, vmin=0, vmax=2*pivot_value)
binvalues = binvalues.astype(np.int)
for i in range(len(x1) - 1):
for j in range(len(y1) - 1):
plt.text((x1[i] + x1[i + 1]) / 2, (y1[j] + y1[j + 1]) / 2, binvalues[i, j],
color='white' if binvalues[i, j] < pivot_value else 'black',
ha='center', va='center', size=8)
plt.show()
PS: If the bin values are very important, you can add them all as ticks. Then, their positions can also be used to draw gridlines as a division between the cells.
plt.yticks(y1)
plt.xticks(x1, rotation=90)
plt.grid(True, ls='-', lw=1, color='black')
To obtain contours based on these data, you could plt.contourf with the generated matrix. (You might want to use np.histogram2d to directly create the matrix.)
plt.contourf((x1[1:]+x1[:-1])/2, (y1[1:]+y1[:-1])/2, binvalues.T, levels=[0,100,1000], cmap=cmap)

Matplotlib legend makes the image too large

I'm plotting this figure with matplotlib, the for loop just color the background:
fig, ax = plt.subplots()
ax.set_ylabel('Number of contacts')
ax.set_xlabel('Time [s]')
for m in range(len(data[node])):
if data[node][m] == -1:
ax.axvline(m,color='r',linewidth=5,alpha=0.2,label="OUT")
if data[node][m] == 0:
ax.axvline(m,color='g',linewidth=5,alpha=0.2,label="RZ0")
if data[node][m] == 1:
ax.axvline(m,color='y',linewidth=5,alpha=0.2,label="RZ1")
ax.plot(x, y, 'b+')
# ax.legend() # HERE is the problem
plt.show()
Which plots the following:
What I want now is a legend to indicate each color of the background meaning, but when I include ax.legend() I get the following error:
ValueError: Image size of 392x648007 pixels is too large. It must be less than 2^16 in each
direction.
<Figure size 432x288 with 1 Axes>
<Figure size 432x288 with 0 Axes>
How am I supposed to name each color of the background, there are 43200 vertical lines but only 3 colors, does it have anything to do with the number of lines?
The trick is to set the label only once. You can add a variable for each label and replace it with None once it's used. Note that using axvline to draw a background has the problem that the line width is measured in pixel space, so neighboring lines will either overlap or have a small white space inbetween. Better to use axvspan. To avoid the white space at the left and at the right, you can explicitly set the x-limits.
The code can be simplified somewhat using a loop.
Updated code:
group consecutive spans together for drawing
precalculate the effect of alpha so the background can be drawn without the need for transparency
from matplotlib import pyplot as plt
from matplotlib import colors as mcolors
import numpy as np
import pandas as pd
import itertools
fig, ax = plt.subplots()
# create some random data
x = np.arange(100)
y = np.sinh(x/20)
indicators = [-1, 0, 1]
node = 0
data = [np.random.choice(indicators, len(x), p=[10/16,1/16,5/16])]
labels = ["OUT", "RZ0", "RZ1"]
colors = ['lime', 'purple', 'gold']
alpha = 0.4
# precalculate the effect of alpha so the colors can be applied with alpha=1
colors = [[1 + (x - 1) * alpha for x in mcolors.to_rgb(c)] for c in colors]
m = 0
for val, group in itertools.groupby(data[node]):
width = len(list(group))
ind = indicators.index(val)
ax.axvspan(m, m + width, color=colors[ind], linewidth=0, alpha=1, label=labels[ind])
labels[ind] = None # reset the label to make sure it is only used once
m += width
ax.plot(x, y, 'b+')
ax.set_xlim(0, len(data[node]))
ax.legend(framealpha=1) # to make the legend background opaque
plt.show()
Do something like hrz1 = ax.axvline(m,color='y',linewidth=5,alpha=0.2) for each of your classes, and then ax.legend((hrz1, hrz0, hout), ('RZ1', 'RZ0', 'OUT'). The hrz1 pointer will be rewritten for each line you make, and then legend will only make one label for each of the handles.

Python Colormap for displaying difference in the high value range

Is there any colormap which displays the differences in high value ranges of an image good, and if not(because I found no cm) is there a way to define it by msyself and give it as argument cmap to seaborn/matplotlib plots? I already tried logarithmic colorscaling but it is hard to define that only the highest 10 percent of your image should be clearly has a nice color difference and the rest could be black for example.
So I have a image/array which has numbers between 0 and 2000. and I only what to clearly differ between numbers between 1800 and 2000.
Is there a way to do that?
Using imshow's vmin and vmax arguments you can restrict the color range to 1800 to 2000. If you want the lower values to be mapped to black you can use cmap.set_under(black).
import numpy as np
import matplotlib.pyplot as plt
t = np.linspace(-7,7,301)
x,y = np.meshgrid(t,t)
z1 = np.sin(x)*np.cos(6*y) + np.cos(12*x*y)/6
z2 = np.exp(-((x-2)**2+(y-2)**2)*3)
z = (z1*6/7+1)*100+1800-z2*1949
cmap=plt.get_cmap("viridis")
cmap.set_under("black")
fig, (ax1, ax2) = plt.subplots(ncols=2, figsize=(9,3.0))
im1 = ax1.imshow(z, cmap=cmap)
fig.colorbar(im1, ax=ax1)
im2 = ax2.imshow(z, vmin=1800, vmax=2000, cmap=cmap)
fig.colorbar(im2, ax=ax2)
ax1.set_title("Original")
ax2.set_title("Linear between 1800 and 2000")
plt.show()
It's not exactly what you want, but you could try to just plot the two groups separately, I mean:
vals = np.array([1,2,3,101,120,150])
xs = np.arange(6)
ys = np.arange(6)
mask = vals>100
vals2 = vals[mask]
xs1 = xs[mask]
xs2 = xs[np.logical_not(mask)]
ys1 = ys[mask]
ys2 = ys[np.logical_not(mask)]
plt.scatter(xs2,ys2,c='black')
plt.scatter(xs1,ys1,c=vals2)
plt.show()
Of course you can adjust the colormap for second plot so it'll differentiate the 'interesting' points from others

Add horizontal line with conditional coloring

I make a contourf plot using matplotlib.pyplot. Now I want to have a horizontal line (or something like ax.vspan would work too) with conditional coloring at y = 0. I will show you what I have and what I would like to get. I want to do this with an array, let's say landsurface that represents either land, ocean or ice. This array is filled with 1 (land), 2 (ocean) or 3 (ice) and has the len(locs) (so the x-axis).
This is the plot code:
plt.figure()
ax=plt.axes()
clev=np.arange(0.,50.,.5)
plt.contourf(locs,height-surfaceheight,var,clev,extend='max')
plt.xlabel('Location')
plt.ylabel('Height above ground level [m]')
cbar = plt.colorbar()
cbar.ax.set_ylabel('o3 mixing ratio [ppb]')
plt.show()
This is what I have so far:
This is what I want:
Many thanks in advance!
Intro
I'm going to use a line collection .
Because I have not your original data, I faked some data using a simple sine curve and plotting on the baseline the color codes corresponding to small, middle and high values of the curve
Code
Usual boilerplate, we need to explicitly import LineCollection
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.collections import LineCollection
Just to plot something, a sine curve (x r
x = np.linspace(0, 50, 101)
y = np.sin(0.3*x)
The color coding from the curve values (corresponding to your surface types) to the LineCollection colors, note that LineCollection requires that the colors are specified as RGBA tuples but I have seen examples using color strings, bah!
# 1 when near min, 2 when near 0, 3 when near max
z = np.where(y<-0.5, 1, np.where(y<+0.5, 2, 3))
col_d = {1:(0.4, 0.4, 1.0, 1), # blue, near min
2:(0.4, 1.0, 0.4, 1), # green, near zero
3:(1.0, 0.4, 0.4, 1)} # red, near max
# prepare the list of colors
colors = [col_d[n] for n in z]
In a line collection we need a sequence of segments, here I have decided to place my coded line at y=0 but you can just add a constant to s to move it up and down.
I admit that forming the sequence of segments is a bit tricky...
# build the sequence of segments
s = np.zeros(101)
segments=np.array(list(zip(zip(x,x[1:]),zip(s,s[1:])))).transpose((0,2,1))
# and fill the LineCollection
lc = LineCollection(segments, colors=colors, linewidths=5,
antialiaseds=0, # to prevent artifacts between lines
zorder=3 # to force drawing over the curve) lc = LineCollection(segments, colors=colors, linewidths=5) # possibly add zorder=...
Finally, we put everything on the canvas
# plot the function and the line collection
fig, ax = plt.subplots()
ax.plot(x,y)
ax.add_collection(lc)
I would suggest adding an imshow() with proper extent, e.g.:
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colorbar as colorbar
import matplotlib.colors as colors
### generate some data
np.random.seed(19680801)
npts = 50
x = np.random.uniform(0, 1, npts)
y = np.random.uniform(0, 1, npts)
X,Y=np.meshgrid(x,y)
z = x * np.exp(-X**2 - Y**2)*100
### create a colormap of three distinct colors for each landmass
landmass_cmap=colors.ListedColormap(["b","r","g"])
x_land=np.linspace(0,1,len(x)) ## this should be scaled to your "location"
## generate some fake landmass types (either 0, 1, or 2) with probabilites
y_land=np.random.choice(3, len(x), p=[0.1, 0.6, 0.3])
print(y_land)
fig=plt.figure()
ax=plt.axes()
clev=np.arange(0.,50.,.5)
## adjust the "height" of the landmass
x0,x1=0,1
y0,y1=0,0.05 ## y1 is the "height" of the landmass
## make sure that you're passing sensible zorder here and in your .contourf()
im = ax.imshow(y_land.reshape((-1,len(x))),cmap=landmass_cmap,zorder=2,extent=(x0,x1,y0,y1))
plt.contourf(x,y,z,clev,extend='max',zorder=1)
ax.set_xlim(0,1)
ax.set_ylim(0,1)
ax.plot()
ax.set_xlabel('Location')
ax.set_ylabel('Height above ground level [m]')
cbar = plt.colorbar()
cbar.ax.set_ylabel('o3 mixing ratio [ppb]')
## add a colorbar for your listed colormap
cax = fig.add_axes([0.2, 0.95, 0.5, 0.02]) # x-position, y-position, x-width, y-height
bounds = [0,1,2,3]
norm = colors.BoundaryNorm(bounds, landmass_cmap.N)
cb2 = colorbar.ColorbarBase(cax, cmap=landmass_cmap,
norm=norm,
boundaries=bounds,
ticks=[0.5,1.5,2.5],
spacing='proportional',
orientation='horizontal')
cb2.ax.set_xticklabels(['sea','land','ice'])
plt.show()
yields:

Setting discrete colormap corresponding to specific data range in Matplotlib

Some background
I have a 2-d array in the shape of (50,50), the data value are range from -40 ~ 40.
But I want to plot the data in three data range[<0], [0,20], [>20]
Then, I need to generate a colormap corresponding to the three section.
I have some thought now
## ratio is the original 2-d array
binlabel = np.zeros_like(ratio)
binlabel[ratio<0] = 1
binlabel[(ratio>0)&(ratio<20)] = 2
binlabel[ratio>20] = 3
def discrete_cmap(N, base_cmap=None):
base = plt.cm.get_cmap(base_cmap)
color_list = base(np.linspace(0, 1, N))
cmap_name = base.name + str(N)
return base.from_list(cmap_name, color_list, N)
fig = plt.figure()
ax = plt.gca()
plt.pcolormesh(binlabel, cmap = discrete_cmap(3, 'jet'))
divider = make_axes_locatable(ax)
cax = divider.append_axes("bottom", size="4%", pad=0.45)
cbar = plt.colorbar(ratio_plot, cax=cax, orientation="horizontal")
labels = [1.35,2,2.65]
loc = labels
cbar.set_ticks(loc)
cbar.ax.set_xticklabels(['< 0', '0~20', '>20'])
Is there any better approach? Any advice would be appreciate.
There are various answers to other questions using ListedColormap and BoundaryNorm, but here's an alternative. I've ignored the placement of your colorbar, as that's not relevant to your question.
You can replace your binlabel calculation with a call to np.digitize() and replace your discrete_cmap() function by using the lut argument to get_cmap(). Also, I find it easier to place the color bounds at .5 midpoints between the indexes rather than scale to awkward fractions of odd numbers:
import matplotlib.colors as mcol
import matplotlib.cm as cm
import matplotlib.pyplot as plt
import numpy as np
ratio = np.random.random((50,50)) * 50.0 - 20.0
fig2, ax2 = plt.subplots(figsize=(5,5))
# Turn the data into an array of N bin indexes (i.e., 0, 1 and 2).
bounds = [0,20]
iratio = np.digitize(ratio.flat,bounds).reshape(ratio.shape)
# Create a colormap containing N colors and a Normalizer that defines where
# the boundaries of the colors should be relative to the indexes (i.e., -0.5,
# 0.5, 1.5, 2.5).
cmap = cm.get_cmap("jet",lut=len(bounds)+1)
cmap_bounds = np.arange(len(bounds)+2) - 0.5
norm = mcol.BoundaryNorm(cmap_bounds,cmap.N)
# Plot using the colormap and the Normalizer.
ratio_plot = plt.pcolormesh(iratio,cmap=cmap,norm=norm)
cbar = plt.colorbar(ratio_plot,ticks=[0,1,2],orientation="horizontal")
cbar.set_ticklabels(["< 0","0~20",">20"])

Categories

Resources