I am trying to understand how to use interactivity in RadioButtonGroup using Bokeh and CustomJS with a Python function. I have tweaked the example provided at the Bokeh site for plotting y=x^f. Instead of using a slider for the power f, I would like to toggle between two powers, f=0.5 and f=0.2. I have followed the manual and inserted RadioButtonGroup in my code using Jupyter notebook. The buttons are showing and responsive, but I am unable to get any callback response out of toggling the buttons.
Any help will be appreciated.
from math import pi
from bokeh.io import output_file, show
from bokeh.layouts import column
from bokeh.models import ColumnDataSource, CustomJS, Slider, TextInput, RadioButtonGroup
from bokeh.plotting import Figure, output_notebook
output_notebook()
x = [x*0.005 for x in range(0, 200)]
y = x
source = ColumnDataSource(data=dict(x=x, y=y))
plot = Figure(plot_width=400, plot_height=400)
plot.line('x', 'y', source=source, line_width=3, line_alpha=0.6)
def callback(source=source, input1=input1, window=None):
data = source.data
m= input1.active
if m==0:
f=.5
else:
f=2
x, y = data['x'], data['y']
for i in range(len(x)):
y[i] = window.Math.pow(x[i], f)
source.trigger('change')
input1 = RadioButtonGroup(labels=["power = .5", "power = 2."], active=0)
input1.button_type="success"
input1.js_on_change('active', CustomJS.from_py_func(callback))
layout = column(input1, plot)
show(layout)
I can not make bokeh.models.RadioButtonGroup generate a callback, but with bokeh.models.RadioGroup I can (I'm using bokeh version 0.12.4 Maybe a newer version facilitates RadioButtonGroup to generate a callback). Also, you are referencing input1 before declaring it. It's not needed as input of your callback function, inside you can use cb_obj. See:
import bokeh
import bokeh.plotting
bokeh.io.output_notebook()
x = [x*0.005 for x in range(0, 200)]
y = x
source = bokeh.models.ColumnDataSource(data=dict(x=x, y=y))
plot = bokeh.plotting.figure(plot_width=400, plot_height=400)
plot.line('x', 'y', source=source, line_width=3, line_alpha=0.6)
def callback(source=source, window=None):
data = source.data
m = cb_obj.active
if m==0:
f=.5
else:
f=2
x, y = data['x'], data['y']
for i in range(len(x)):
y[i] = window.Math.pow(x[i], f)
source.trigger('change')
input1 = bokeh.models.RadioGroup(labels=["power = .5", "power = 2."],active=0,
callback=bokeh.models.CustomJS.from_py_func(callback))
layout = bokeh.layouts.column(input1, plot)
bokeh.io.show(layout)
Related
I would like to add a separate colorbar to each subplot in a 2x2 plot.
fig , ( (ax1,ax2) , (ax3,ax4)) = plt.subplots(2, 2,sharex = True,sharey=True)
z1_plot = ax1.scatter(x,y,c = z1,vmin=0.0,vmax=0.4)
plt.colorbar(z1_plot,cax=ax1)
z2_plot = ax2.scatter(x,y,c = z2,vmin=0.0,vmax=40)
plt.colorbar(z1_plot,cax=ax2)
z3_plot = ax3.scatter(x,y,c = z3,vmin=0.0,vmax=894)
plt.colorbar(z1_plot,cax=ax3)
z4_plot = ax4.scatter(x,y,c = z4,vmin=0.0,vmax=234324)
plt.colorbar(z1_plot,cax=ax4)
plt.show()
I thought that this is how you do it, but the resulting plot is really messed up; it just has an all grey background and ignores the set_xlim , set_ylim commands I have (not shown here for simplicity). + it shows no color bars. Is this the right way to do it?
I also tried getting rid of the "cax = ...", but then the colorbar all goes on the bottom right plot and not to each separate plot!
This can be easily solved with the the utility make_axes_locatable. I provide a minimal example that shows how this works and should be readily adaptable:
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable
import numpy as np
m1 = np.random.rand(3, 3)
m2 = np.arange(0, 3*3, 1).reshape((3, 3))
fig = plt.figure(figsize=(16, 12))
ax1 = fig.add_subplot(121)
im1 = ax1.imshow(m1, interpolation='None')
divider = make_axes_locatable(ax1)
cax = divider.append_axes('right', size='5%', pad=0.05)
fig.colorbar(im1, cax=cax, orientation='vertical')
ax2 = fig.add_subplot(122)
im2 = ax2.imshow(m2, interpolation='None')
divider = make_axes_locatable(ax2)
cax = divider.append_axes('right', size='5%', pad=0.05)
fig.colorbar(im2, cax=cax, orientation='vertical');
In plt.colorbar(z1_plot,cax=ax1), use ax= instead of cax=, i.e. plt.colorbar(z1_plot,ax=ax1)
Specify the ax argument to matplotlib.pyplot.colorbar(), e.g.
import numpy as np
import matplotlib.pyplot as plt
fig, ax = plt.subplots(2, 2)
for i in range(2):
for j in range(2):
data = np.array([[i, j], [i+0.5, j+0.5]])
im = ax[i, j].imshow(data)
plt.colorbar(im, ax=ax[i, j])
plt.show()
Please have a look at this matplotlib example page. There it is shown how to get the following plot with four individual colorbars for each subplot:
I hope this helps.
You can further have a look here, where you can find a lot of what you can do with matplotlib.
Try to use the func below to add colorbar:
def add_colorbar(mappable):
from mpl_toolkits.axes_grid1 import make_axes_locatable
import matplotlib.pyplot as plt
last_axes = plt.gca()
ax = mappable.axes
fig = ax.figure
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="5%", pad=0.05)
cbar = fig.colorbar(mappable, cax=cax)
plt.sca(last_axes)
return cbar
Then you codes need to be modified as:
fig , ( (ax1,ax2) , (ax3,ax4)) = plt.subplots(2, 2,sharex = True,sharey=True)
z1_plot = ax1.scatter(x,y,c = z1,vmin=0.0,vmax=0.4)
add_colorbar(z1_plot)
I am new to Matplotlib. Based on my code in following, I wanted to update the data,title,xlabel,ylabel at same time. However, the title and labels did not been updated, but data did.Someone can give me a solution? That will help me a lot.Thank you.
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
def updata(frame_number):
current_index = frame_number % 3
a = [[1,2,3],[4,5,6],[7,8,9]]
idata['position'][:,0] = np.asarray(a[current_index])
idata['position'][:,1] = np.asarray(a[current_index])
scat.set_offsets(idata['position'])
ax.set_xlabel('The Intensity of Image1')
ax.set_ylabel('The Intensity of Image2')
ax.set_title("For Dataset %d" % current_index)
fig = plt.figure(figsize=(5,5))
ax = fig.add_axes([0,0,1,1])
idata = np.zeros(3,dtype=[('position',float,2)])
ax.set_title(label='lets begin',fontdict = {'fontsize':12},loc='center')
scat = ax.scatter(idata['position'][:,0],idata['position'][:,1],s=10,alpha=0.3,edgecolors='none')
animation = FuncAnimation(fig,updata,interval=2000)
plt.show()
Running the code, I see an empty window. The reason is that the axes span the complete figure (fig.add_axes([0,0,1,1])). In order to see the title and labels, you would need to make the axes smaller than the figure, e.g. by
ax = fig.add_subplot(111)
Also, the scale of the axes is not defined, so the animation will happen outside the axes limits. You can use ax.set_xlim and ax.set_ylim to prevent that.
Here is a complete running code:
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
def updata(frame_number):
current_index = frame_number % 3
a = [[1,2,3],[4,5,6],[7,8,9]]
idata['position'][:,0] = np.asarray(a[current_index])
idata['position'][:,1] = np.asarray(a[current_index])
scat.set_offsets(idata['position'])
ax.set_xlabel('The Intensity of Image1')
ax.set_ylabel('The Intensity of Image2')
ax.set_title("For Dataset %d" % current_index)
fig = plt.figure(figsize=(5,5))
ax = fig.add_subplot(111)
idata = np.zeros(3,dtype=[('position',float,2)])
ax.set_title(label='lets begin',fontdict = {'fontsize':12},loc='center')
scat = ax.scatter(idata['position'][:,0],idata['position'][:,1],
s=25,alpha=0.9,edgecolors='none')
ax.set_xlim(0,10)
ax.set_ylim(0,10)
animation = FuncAnimation(fig,updata,frames=50,interval=600)
plt.show()
I have written a program that creates a graph based on input from a dropdown list. I am using curdoc().add_root() from bokeh to show my graphs on a server as show() does not work. However, whenever I choose a new option, instead of replacing the current graph, it creates one below it. I have tried curdoc().clear() its not working. How do I make this work where it replaces the graph but doesnt delete the dropdown list, because that is what curdoc().clear() is doing? Here's my code:
import csv
import bokeh.plotting
from bokeh.plotting import figure, curdoc
from bokeh.io import output_file, show
from bokeh.layouts import widgetbox
from bokeh.models.widgets import MultiSelect
from bokeh.io import output_file, show, vform
from bokeh.layouts import row
from collections import defaultdict
columns = defaultdict(list) # each value in each column is appended to a list
columns1 = defaultdict(list)
with open('my_data.csv') as f:
for row in f:
row = row.strip()# read a row as {column1: value1, column2: value2,...}
row = row.split(',')
columns[row[0]].append(row[1])
columns[row[0]].append(row[2])
columns[row[0]].append(row[3])
columns[row[0]].append(row[4])
columns[row[0]].append(row[5])
with open('my_data1.csv') as f:
for row in f:
row = row.strip()# read a row as {column1: value1, column2: value2,...}
row = row.split(',')
columns1[row[0]].append(row[1])
columns1[row[0]].append(row[2])
columns1[row[0]].append(row[3])
columns1[row[0]].append(row[4])
columns1[row[0]].append(row[5])
from bokeh.layouts import widgetbox
from bokeh.models.widgets import Dropdown
from bokeh.plotting import curdoc
menu = [("NY", "New York"), ("California", "California"), ("Ohio", "Ohio")]
dropdown = Dropdown(label="Dropdown button", button_type="warning", menu=menu)
count = 0
#def function_to_call(attr, old, new):
#print dropdown.value
def myfunc(attr, old, new):
aaa = dropdown.value
xy = (columns[aaa])
xy = [float(i) for i in xy]
myInt = 10000
xy = [x / myInt for x in xy]
print xy
omega = (columns1[aaa])
omega = [float(i) for i in omega]
print omega
import numpy
corr123 = numpy.corrcoef(omega,xy)
print corr123
a = [2004, 2005, 2006, 2007, 2008]
p = figure(tools="pan,box_zoom,reset,save", title="Diabetes and Stats",
x_axis_label='Years', y_axis_label='percents')
# add some renderers
per = "Diabetes% " + aaa
p.line(a, omega, legend=per)
p.circle(a, omega, legend=per, fill_color="white",line_color="green", size=8)
p.line(a, xy, legend="Per Capita Income/10000")
p.circle(a, xy, legend="Per Capita Income/10000", fill_color="red", line_color="red", size=8)
p.legend.location="top_left"
#bokeh.plotting.reset_output
#curdoc().clear()
curdoc().add_root(p)
curdoc().add_root(dropdown)
#bokeh.plotting.reset_output
dropdown.on_change('value', myfunc)
curdoc().add_root(dropdown)
In a scatter plot matrix, I would like to draw a region in every subplot and print the points that are included in the region. I found the LassoSelector widget, which does exactly that. I am trying to extend its functionality for more than one subplots. I am getting the following error: self.xys = collection.get_offsets(),
AttributeError: 'numpy.flatiter' object has no attribute 'get_offsets'.
when the line selector = SelectFromCollection(axes, ax.flat) is in the for loop, and I am getting the error: self.canvas = ax.figure.canvas,AttributeError: 'numpy.ndarray' object has no attribute 'figure' when the line selector = SelectFromCollection(ax, ax.flat) is outside of the loop. Why does this happen?
Here is my code:
from __future__ import print_function
import numpy as np
from matplotlib.widgets import LassoSelector
from matplotlib.path import Path
class SelectFromCollection(object):
"""Select indices from a matplotlib collection using `LassoSelector`.
Selected indices are saved in the `ind` attribute. This tool highlights
selected points by fading them out (i.e., reducing their alpha values).
If your collection has alpha < 1, this tool will permanently alter them.
Note that this tool selects collection objects based on their *origins*
(i.e., `offsets`).
Parameters
----------
ax : :class:`~matplotlib.axes.Axes`
Axes to interact with.
collection : :class:`matplotlib.collections.Collection` subclass
Collection you want to select from.
alpha_other : 0 <= float <= 1
To highlight a selection, this tool sets all selected points to an
alpha value of 1 and non-selected points to `alpha_other`.
"""
def __init__(self, ax, collection, alpha_other=0.3):
self.canvas = ax.figure.canvas
self.collection = collection
self.alpha_other = alpha_other
self.xys = collection.get_offsets()
self.Npts = len(self.xys)
# Ensure that we have separate colors for each object
self.fc = collection.get_facecolors()
if len(self.fc) == 0:
raise ValueError('Collection must have a facecolor')
elif len(self.fc) == 1:
self.fc = np.tile(self.fc, self.Npts).reshape(self.Npts, -1)
self.lasso = LassoSelector(ax, onselect=self.onselect)
self.ind = []
def onselect(self, verts):
path = Path(verts)
self.ind = np.nonzero([path.contains_point(xy) for xy in self.xys])[0]
self.fc[:, -1] = self.alpha_other
self.fc[self.ind, -1] = 1
self.collection.set_facecolors(self.fc)
self.canvas.draw_idle()
print(selector.xys[selector.ind])
#selector.disconnect()
def disconnect(self):
self.lasso.disconnect_events()
self.fc[:, -1] = 1
self.collection.set_facecolors(self.fc)
self.canvas.draw_idle()
if __name__ == '__main__':
import matplotlib.pyplot as plt
plt.ion()
data=np.loadtxt(r"data.txt")
x = data[:, 3]
x1 = data[:, 4]
y = data[:,5]
y1 = data[:,6]
fig, ax = plt.subplots(nrows=2, ncols=2, squeeze=True)
for axes, marker in zip(ax.flat, ['o', 'o']):
ax.flat[0].plot(x, y, 'r', ls='', marker=marker)
ax.flat[1].plot(x, x1,'r', ls='', marker=marker)
ax.flat[2].plot(x, y1,'r', ls='', marker=marker)
ax.flat[3].plot(y, x1,'r', ls='', marker=marker)
selector = SelectFromCollection(ax, ax.flat)
plt.show(block=True)
plt.draw()
Ok, I found a few problems that are causing your code not to work properly. There we go:
Firts of all, you modified the SelectFromCollection class that you got from the LassoSelector example to print every selected point, but forgot a detail:
class SelectFromCollection(object):
def __init__(self, ax, collection, alpha_other=0.3):
# ...
# No changes here...
# ...
def onselect(self, verts):
path = Path(verts)
self.ind = np.nonzero([path.contains_point(xy) for xy in self.xys])[0]
self.fc[:, -1] = self.alpha_other
self.fc[self.ind, -1] = 1
self.collection.set_facecolors(self.fc)
self.canvas.draw_idle()
print(self.xys[self.ind]) # <- THIS LINE HAS CHANGED!!!
#selector.disconnect()
def disconnect(self):
# ...
# No changes here...
# ...
Now you can use multiple instances of SelectFromCollection.
Then, you are also creating only one instance of SelectFromCollection (so only one subplot would react). Furthermore, according to the doctsring the second argument the __init__ method expects is a matplotlib.collections.Collection instance.
Instead you are passing it a numpy array (in fact a numpy.Flatiter) that contains two Axes instances. If you look at the example, there it gets a Collection instance returned by the scattercommand (they use scatter instead of plot).
All in all, and restiling the loop, this is my version
if __name__ == '__main__':
import matplotlib.pyplot as plt
data=np.random.rand(3,100)
xdata = data[:-1] # all rows but last
y = data[-1] # last row
fig, axes = plt.subplots(nrows=1, ncols=2, squeeze=True)
markers = ['o','^']
selectors =[]
for i in xrange(xdata.shape[0]):
pts = axes[i].scatter(xdata[i], y, c='r', marker=markers[i])
selectors.append(SelectFromCollection(axes[i], pts))
plt.show()
EDIT
If you want to do more plots, it is not hard. You can try to write more synthetic code with a for loop and so on, but an easier solution is to write directly the repetitions of the code:
if __name__ == '__main__':
import matplotlib.pyplot as plt
data=np.loadtxt(r"data.txt")
x = data[:, 3]
x1 = data[:, 4]
y = data[:,5]
y1 = data[:,6]
fig, axes = plt.subplots(nrows=2, ncols=2)
pts1 = axes[0,0].scatter(x, y, c='r', marker='o')
select1 = SelectFromCollection(axes[0,0], pts1)
pts2 = axes[1,0].scatter(x, x1, c='r', marker='o')
select2 = SelectFromCollection(axes[1,0], pts2)
pts3 = axes[0,1].scatter(x, y1, c='r', marker='o')
select3 = SelectFromCollection(axes[0,1], pts3)
pts4 = axes[1,1].scatter(y, x1, c='r', marker='o')
select4 = SelectFromCollection(axes[1,1], pts4)
plt.show()
Still, it is necessary that you change the definition of the SelectFromCollection class as I said above.
i'm using matplotlib with django. I'm trying to create bar charts.
i followed the cookbook, but i just got a grey rectangular box.
Now I'm using the following code, and have a title and axes.
How can I add a bar graph to the figure? Currently there is no actual data inside the axes.
Here's my charting code:
from matplotlib.backends.backend_agg import FigureCanvasAgg
from matplotlib.figure import Figure
import matplotlib.pyplot as plt
class Chart(object):
## Creates a bar chart of the given data
#staticmethod
def bar(data):
figure = Figure(figsize=(6,6))
ax = figure.add_axes([0.1, 0.1, 0.8, 0.8])
labels = 'Frogs', 'Hogs', 'Dogs', 'Logs'
fracs = [15, 30, 45, 10]
explode=(0, 0.05, 0, 0)
plt.pie(fracs, explode=explode, labels=labels, autopct='%1.1f%%', shadow=True)
figure.suptitle('Raining Hogs and Dogs', fontsize=14)
canvas = FigureCanvasAgg(figure)
return canvas
In my view I have:
canvas = Chart.bar(results)
# turn the returned canvas into an HTTP response
response=HttpResponse(content_type='image/png')
canvas.print_png(response)
return response
fig = Figure()
fig = Figure(facecolor='white', edgecolor='white')
ax = fig.add_subplot(1,1,1)
x = matplotlib.numpy.arange(0, len(dic.keys()))
ind = matplotlib.numpy.arange(len(dic.values()))
height = 0.8
ax.bar(ind, dic.values(), width, color=colors)
ax.set_xticks(ind + width / 2.0)
ax.set_xticklabels(dic.keys())
padding = 0.2
ax.set_xlim([x.min() - padding, x.max() + width + padding])
canvas = FigureCanvas(fig)
response = django.http.HttpResponse(content_type='image/png')
canvas.print_png(response)
fig.savefig(filename)
this will create a bar graph, and save the image. Just have to call the function into your views. and open the image in the template. I passed a dictionary to this function(dic) but you can pass a list, is up to you.
in this case the keys are the x axis and the values are the y axis.