fix attributeerror in tutorial part10

[Rachelxuan11]

Problem
Run tutorial Part10 and got the same error as this issue #2758.
This error is shown as follows:

in train_enc()
17 spdz_params = list()
18 for remote_index in range(len(compute_node)): # for each client
---> 19 spdz_params.append(params[remote_index][param_i].copy().fix_precision().share(bob, alice, crypto_provider=james).get()) # AttributeError: 'numpy.ndarray' object has no attribute 'wrap', if .get() is used here. This could be avoided by model.get() first, and then .share() the copy to a list of clients. This obeys the actually protocol of encrypted model averaging, but fixes this bug.

Solution
user model.copy().get().fix_precision().share(...) instead of model.copy().fix_precision().share(...).get() could avoid this error

[review-notebook-app]

Check out this pull request on 

You'll be able to see Jupyter notebook diff and discuss changes. Powered by ReviewNB.

[Rachelxuan11]

a small change

[Rachelxuan11]

redundant for this example regression task

[Rachelxuan11]

avoid the AttributeError

[gmuraru]

It seems that there is an error here (it is lalso present on master, but also in this PR). Do you have some time to investigate :)

~/contrib/tests/PySyft/syft/generic/frameworks/hook/hook_args.py in <lambda>(i, **kwargs)
    742         if isinstance(r, (list, tuple))  # if the rule is a list or tuple.
    743         # Last if not, rule is probably == 1 so use type to return the right transformation.
--> 744         else lambda i, **kwargs: register_tensor(i, **kwargs)
    745         for a, r in zip(response, rules)  # And do this for all the responses / rules provided
    746     ]

~/contrib/tests/PySyft/syft/generic/frameworks/hook/hook_args.py in register_tensor(tensor, owner, response_ids)
    710             and each id is pop out when needed.
    711     """
--> 712     tensor.owner = owner
    713     try:
    714         tensor.id = response_ids.pop(-1)

AttributeError: 'numpy.ndarray' object has no attribute 'owner'

[github-actions]

This pull request has been marked stale because it has been open 30 days with no activity. Leave a comment or remove the stale label to unmark it. Otherwise, this will be closed in 7 days.