Create Vignette - Documentation for multi axis figures

[lwasser]

https://github.com/earthlab/autograding-notebooks/blob/master/notebooks/ea-homework-for-timeseries-raster.ipynb

# Plot the data - place your final data here

# BEGIN SOLUTION

# Add plot code only for plots 1 and 2 here
fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(10, 10))
ep.plot_rgb(naip_2017_crop,
            rgb=[0, 1, 2],
            extent=naip_2015_ext,
            title="Homework PLOT 1: NAIP 2017 Post Fire RGB Image\n Cropped",
            ax=ax1)
fire_bound_utmz13.plot(ax=ax1, color='None',
                       edgecolor='white', linewidth=2)

# plot 2017 cropped data
ep.plot_rgb(naip_2017_crop,
            rgb=[3, 0, 1],
            extent=naip_2015_ext,
            title="Homework PLOT 2: NAIP 2017 Post Fire CIR Image",
            ax=ax2)
fire_bound_utmz13.plot(ax=ax2, color='None',
                       edgecolor='white', linewidth=2)
plt.show()
# END SOLUTION

### DO NOT REMOVE LINE BELOW ###
raster_plot = nb.convert_axes(plt, which_axes="all")

# Here i have to create a set of tests for a plot with subplots. this might involve loops
hw_raster_plot1 = ra.RasterTester(raster_plot[0])
hw_raster_plot2 = ra.RasterTester(raster_plot[1])

test_axis_off_plot1 = [hw_raster_plot1.assert_axis_off(),
                       1,
                       "RGB Image: Axis format looks great!",
                       "RGB Image: Looks like your axis is still on. Please turn axis_off."]

test_axis_off_plot2 = [hw_raster_plot2.assert_axis_off(),
                       1,
                       "CIR Image: Axis format looks great!",
                       "CIR Image: Looks like your axis is still on. Please turn axis_off."]

# Check plot extent ?? -- not sure i can do that yet with MPC functionality but it would be easy to add

# Check title
test_title_plot1 = [hw_raster_plot1.assert_title_contains(["NAIP", "2017", "RGB"]),
                    2,
                    "Title looks good",
                    "oops check your title"]
test_title_plot2 = [hw_raster_plot2.assert_title_contains(["NAIP", "2017", "CIR"]),
                    2,
                    "Title looks good",
                    "oops check your title"]

run_all_tests([test_axis_off_plot1, 
               test_axis_off_plot2, 
               test_title_plot1, 
               test_title_plot2])

[nkorinek]

@lwasser for this Vignette, we've been talking recently about just using the ax object instead of using the nb.convert_axes() function. Would you like me to show both, or is there one we want to promote as the "correct" way of checking this?

[lwasser]

i think we want to use the ax approach in our docs. but in each vignette we can add a note to show what the notebook implementation would look like. i think that makes sense for the time being because we can't assume everyone is using notebooks. We could then have an entire vignette dedicated to notebooks

[nkorinek]

Addressed here: https://github.com/earthlab/matplotcheck/pull/241