Decrease batch size for memory optimization (error: interrupted by signal 9: SIGKILL)

[alanjacob10]

Hi,

I'm looking for an extension to the following code:

`from recbole.quick_start import run_recbole

parameter_dict = {
'data_path': "/Users/alanolewnik/Documents/PyCharm/Recommender_new/dataset",
'load_col': {'inter': ['user_id', 'product_id']},
'USER_ID_FIELD': 'user_id',
'ITEM_ID_FIELD': 'product_id'
}

run_recbole(model='GCMC', dataset='steam', config_dict=parameter_dict)`

My goal is to set limiters to the batch size so i have a maximum number of users, items and interactions.
currently I can't run any algorithms because it uses to much memory and my os crashes the program and gives the following error:

Process finished with exit code 137 (interrupted by signal 9: SIGKILL)

Note that this is not me but the os stopping the algorithm.

What should I add to the given code to give a maximum of 100.000 users, and lower item and interaction number also?

Or if you have any other tips on memory optimization for running the algo on MacOS.

Sincerely,
Alan