naszilla / reczilla

RecZilla: Metalearning for algorithm selection on Recommender Systems

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SlopOne hyperparams

duncanmcelfresh opened this issue · comments

It looks like there is no defined hyperparam space for SlopeOne? This should be handled in algorithm_handler.py

Traceback (most recent call last):
  File "/home/shared/miniconda3/envs/reczilla/lib/python3.6/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/home/shared/miniconda3/envs/reczilla/lib/python3.6/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/shared/reczilla/RecSys2019_DeepLearning_Evaluation/Experiment_handler/run_experiment.py", line 117, in <module>
    run(args)
  File "/home/shared/reczilla/RecSys2019_DeepLearning_Evaluation/Experiment_handler/run_experiment.py", line 41, in run
    args.param_seed,
  File "/home/shared/reczilla/RecSys2019_DeepLearning_Evaluation/Experiment_handler/Experiment.py", line 366, in run_experiment
    alg_seed=alg_seed
  File "/home/shared/reczilla/RecSys2019_DeepLearning_Evaluation/ParameterTuning/RandomSearch.py", line 86, in search
    sampler_args=sampler_args,
  File "/home/shared/reczilla/RecSys2019_DeepLearning_Evaluation/ParameterTuning/ParameterSpace.py", line 53, in random_samples
    raise Exception("no hyperparameters. n must be 1")
Exception: no hyperparameters. n must be 1

if SlopeOne is meant to have no hyperparams, then we can close the issue.

It does not have any.