I find the problem can be solved in this way.
Still Thanks!

Is there a reason you can't set s to a different value, like s=10000?

For future reference: Only marker positions are included in auto-scaling; marker sizes are not. For really large markers this can lead to observed clipping.

Whether exists a more convenient way to scatter with a given radius? (suppose it's a circle marker)

Is there a reason you can't set s to a different value, like s=10000?

Actually a different value can be set.

@GYHHAHA Depending on what you are doing, you may be better off using https://matplotlib.org/api/_as_gen/matplotlib.patches.Circle.html It is a bit fussier to use (you will have to create the artists and add them to the axes your self), but it will let you control the size of the circles in data space (rather than in pixel space).

An alternative workaround is to impose larger margins around the data: plt.rcParams['axes.ymargin'] = 0.2 when you know, you have very large markers.

Indeed, it's more convenient to apply circle under this situation. Thanks. @tacaswell

In addition to the options shown in the SO post OP linked to, probably the fastest option would be to use a PathCollection.

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.collections as mcoll
import matplotlib.transforms as mtrans
import matplotlib.markers as mmarkers

c = mcoll.PathCollection(
                [mmarkers.MarkerStyle("o").get_path()],
                offsets=np.repeat(np.arange(10),2).reshape(10,2),
                edgecolors=["k"]*10,
                transOffset=mtrans.IdentityTransform(),
                offset_position="data")
plt.gca().add_collection(c)
plt.gca().set(xlim=(-2, 11), ylim=(-2,11))
plt.gca().set_aspect("equal")
plt.show()

I think in view of this, it's relevant to discuss the possibility to either not deprecate offset_position="data" or to provide a public replacement option.

Yes, it seems more natural. Thanks for your solution! @ImportanceOfBeingErnest