"""
====================================
Creating Virtual Images from Vectors
====================================

In this example we will show how to create virtual images from vectors.
We will use the `find_peaks` method to find the vectors and then use the
`get_unique_vectors` method to reduce the number of vectors to a unique set of vectors.

We can convert the unique vectors to regions of interest (ROIs) objects using the `to_roi` method
and then use the `get_virtual_image` method to create the virtual images.

This method is flexible and any type of ROI defined in hyperspy can be used to create virtual images.

We also included the include_labels argument to the `to_roi` method which will also return the labels
for each ROI for use plotting on the signal.
"""

import pyxem as pxm

c = pxm.data.dummy_data.get_cbed_signal()
pks = c.template_match_disk(5, subtract_min=False).find_peaks(
    interactive=False, threshold_abs=0.8, min_distance=6
)
dv = pxm.signals.DiffractionVectors.from_peaks(pks)
vectors = dv.get_unique_vectors()
roi, texts = vectors.to_roi(radius=5, include_labels=True, sizes=3, facecolor="black")

c.plot()
for r in roi:
    r.add_widget(c, axes=(2, 3))

vdfs = c.get_virtual_image(roi)
c.add_marker(texts)

# %%

vdfs.plot()
# %%
# sphinx_gallery_thumbnail_number = 2