fix: Change formats of user_ids and item_ids arg.

[ywkim92]

Description

ML model predicts the rating for an user inside prepare_all_prediction function.
But if I run the current code, the following error occurs:
TypeError: Invalid type passed to user_ids parameter. This must be either int or np.int32 array. Type received: <class 'numpy.int64'>
To fix this error,

1. the types of user_idslink should be changed from int to np.int32 array
2. the types of item_idslink should be changed from list to np.int32 array

Related Issues

I haven't found yet.

Checklist:

• I have followed the contribution guidelines and code style for this project.
• I have added tests covering my contributions.
• I have updated the documentation accordingly.
• This PR is being made to staging branch and not to main branch.

[gramhagen]

should we add the dtype to ensure it's np.int32?
e.g.: np.array(x["uid"], dtype=np.int32)

[ywkim92]

should we add the dtype to ensure it's np.int32? e.g.: np.array(x["uid"], dtype=np.int32)

oh i forgot it. Thank you so much!

[miguelgfierro]

@ywkim92 this is great, thanks, however there is a conflict in the code so we can't merge, can you please review?

[ywkim92]

@ywkim92 this is great, thanks, however there is a conflict in the code so we can't merge, can you please review?

Hello, thanks for your kind comments. I resolved the conflict as you said. Could you please check this issue again and git me some hints if the conflict is still not resolved?

[SafetyMary]

May i know if the issue has been fixed? If not, is there any work around?
I tried running the lightFM quick start notebook (https://github.com/microsoft/recommenders/blob/main/examples/02_model_hybrid/lightfm_deep_dive.ipynb)
and got the error on

with Timer() as test_time:
all_predictions = prepare_all_predictions(data, uid_map, iid_map,
interactions=train_interactions,
model=model1,
num_threads=NO_THREADS)

TypeError: Invalid type passed to user_ids parameter. This must be either int or np.int32 array. Type received: <class 'numpy.int64'>

System version: 3.8.13 (default, Mar 28 2022, 06:59:08) [MSC v.1916 64 bit (AMD64)]
LightFM version: 1.16

[miguelgfierro]

SafetyMary if you install recommenders from main, the code in this PR should be in. We will do a release in PyPI soon that will include this

[SafetyMary]

Have updated to the lastest 1.1.0 ver through pip but the issue still presists.