`alpha` argument of `ElasticTransform()` should completely avoid negative values, giving error and the doc should have the explanation.

📚 The doc issue

Setting 1000 and -1000 to alpha argument of ElasticTransform() gets the same kind of results as shown below:

from torchvision.datasets import OxfordIIITPet
from torchvision.transforms.v2 import ElasticTransform

my_data = OxfordIIITPet(
    root="data"
)

import matplotlib.pyplot as plt

def show_images2(data, main_title=None, a=50, s=5, f=0):
    plt.figure(figsize=(10, 5))
    plt.suptitle(t=main_title, y=0.8, fontsize=14)
    for i, (im, _) in zip(range(1, 6), data):
        plt.subplot(1, 5, i)
        et = ElasticTransform(alpha=a, sigma=s, fill=f) # Here
        plt.imshow(X=et(im)) # Here
        plt.xticks(ticks=[])
        plt.yticks(ticks=[])
    plt.tight_layout()
    plt.show()

show_images2(data=my_data, main_title="alpha1000_data", a=1000) # Here
show_images2(data=my_data, main_title="alpha-1000_data", a=-1000) # Here

Suggest a potential alternative/fix

So, alphe argument should completely avoid negative values, giving error and the doc should have the explanation.

Start at the torchvision.transforms.v2.ElasticTransform entry point and reproduce the documented example with positive and negative alpha values. Update the behavior so negative alpha is rejected, then update the linked ElasticTransform documentation to explain the constraint and verify the example no longer produces equivalent results.