so what's the usage of trainable_wrap in dynamic_embedding.embedding_lookup?

[andrew-zzz]

does that mean the variable of these ids? but why the values of team are different between c and d

Code to reproduce the issue

import tensorflow_recommenders_addons as tfra
user_id_val = np.array([1,2,3])
c,d = tfra.dynamic_embedding.embedding_lookup(
params=a,
ids=user_id_val,
name="user-id-weights",
return_trainable=True)
print(c)
print('--------')
print(d)

Other info / logs

Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.