Retrieving data from pyplot using geopandas

Question

I'm trying to avoid saving a matplotlib figure to file, just to recall it and preprocess the image for a tensorflow model. I saw this which made me optimistic, unfortunately, geopandas has an inbuilt plotting function. I am wondering if anyone has any ideas?

Ideally, I would like a numPy image array of the intensity values of all points

data = gp.read_file('/path_to.shp')

fig = plt.figure()
ax = fig.add_subplot(111)

lines = data.plot(ax=ax)
print(lines[0].get_data())

Unfortunately, I get the error:

Traceback (most recent call last):
  File "GIS2img.py", line 108, in <module>
    print(lines[0].get_data())
TypeError: 'AxesSubplot' object does not support indexing

I would like to get RGB image data in an array for each figure

I would appreciate any help, thanks

Edit: Clarified question

Text and code are somehow contradictory. It's hence not clear if you want to get the data, or if you want an RGB array of the complete axes. — ImportanceOfBeingErnest, May 02 '19 at 11:50
Ideally, I'd like an RGB array of the complete axes. Will update question — Robbie Sewell, May 02 '19 at 20:20
See, e.g. [this example](https://matplotlib.org/gallery/misc/agg_buffer_to_array.html). — ImportanceOfBeingErnest, May 02 '19 at 20:23

score 0 · Accepted Answer · answered May 02 '19 at 20:46

0

Solved!

Thanks to ImportanceOfBeingErnest for the help.

You can draw to a canvas using: fig.canvas.draw() and then retrieve that image using np.array(fig.canvas.renderer._renderer).

answered May 02 '19 at 20:46

Robbie Sewell

11
3

Retrieving data from pyplot using geopandas

1 Answers1