Unique Legend Image and Line

[nettesheim2]

I would like to plot 2 curves and, in the legend, have a unique image for each curve and the corresponding curve color to show which it belongs to.  I cannot figure out how to replace the text "Curve 1" or "Curve 2" with a sample of the actual curve, as you would usually see in the place of the monkey/giraffe in my case.

I don't have  a great grasp on the handler, but I have enough to create the following code. See the attached image for clarification. To reproduce, you can download a monkey and giraffe png (monkey.png and giraffe.png) into the same directory you run the code in.

So instead of the legend showing:
image1 "curve 1" (where "Curve 1 is the actual text"
image2 "curve 2"

I would like:
image1 curve1 (where curve1 is a segment representing curve 1 (green line)
image2 curve2

I appreciate any help you can offer!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105

import os   from matplotlib.transforms import TransformedBbox from matplotlib.image import BboxImage from matplotlib.legend_handler import HandlerBase from matplotlib._png import read_png from matplotlib.cbook import get_sample_data from matplotlib.transforms import Bbox import numpy as np import matplotlib.pyplot as plt import matplotlib matplotlib.rc('text', usetex = True)     class ImageHandler(HandlerBase):     def create_artists(self, legend, orig_handle,                        xdescent, ydescent, width, height, fontsize,                        trans):           # enlarge the image by these margins         sx, sy = self.image_stretch           # create a bounding box to house the image         bb = Bbox.from_bounds(xdescent - sx,                               ydescent - sy,                               width + sx,                               height + sy)           tbb = TransformedBbox(bb, trans)         image = BboxImage(tbb)         image.set_data(self.image_data)           self.update_prop(image, orig_handle, legend)           return [image]       def set_image(self, image_path, image_stretch=(0, 0)):         if not os.path.exists(image_path):             #sample = get_sample_data("grace_hopper.png", asfileobj=False)             sample = "monkey.png"             self.image_data = read_png(sample)         else:             self.image_data = read_png(image_path)           self.image_stretch = image_stretch class ImageHandler2(HandlerBase):     def create_artists(self, legend, orig_handle,                        xdescent, ydescent, width, height, fontsize,                        trans):           # enlarge the image by these margins         sx, sy = self.image_stretch           # create a bounding box to house the image         bb = Bbox.from_bounds(xdescent - sx,                               ydescent - sy,                               width + sx,                               height + sy)           tbb = TransformedBbox(bb, trans)         image = BboxImage(tbb)         image.set_data(self.image_data)           self.update_prop(image, orig_handle, legend)           return [image]       def set_image(self, image_path, image_stretch=(0, 0)):         if not os.path.exists(image_path):             #sample = get_sample_data("grace_hopper.png", asfileobj=False)             sample = "giraffe.png"             self.image_data = read_png(sample)         else:             self.image_data = read_png(image_path)           self.image_stretch = image_stretch   lw = 1.5;      # LineWidth   # Random curve d1 = np.linspace(25,3500); h1 = 2.020 * d1 **(0.449); h2 = 2.8 * d1 **(0.467); plt.figure() s, = plt.plot(d1,h1,'g',linewidth=lw) s2, = plt.plot(d1,h2,'b',linewidth=lw)   # setup the handler instance for the scattered data custom_handler = ImageHandler() custom_handler.set_image("[PATH TO IMAGE]",                          image_stretch=(0, 20)) # this is for grace hopper custom_handler2 = ImageHandler2() custom_handler2.set_image("[PATH TO IMAGE]",                          image_stretch=(0, 20)) # this is for grace hopper   # add the legend for the scattered data, mapping the # scattered points to the custom handler   plt.legend([s, s2],            ['Curve 1', 'Curve 2'],            handler_map={s: custom_handler, s2: custom_handler2},            labelspacing=2,            frameon=False,            loc=2) plt.show()

[Exponential_Sinusoid]

To be clear, you want to use the plotted function as the thumbnail for that plot alone? If I recall, with matplotlib, you should be able to generate an image of each plot and save them as the aforementioned "giraffe" and "monkey", no?