Open zuoxu3310 opened 3 months ago
Hi @zuoxu3310 , have you tried https://github.com/microsoft/FLAML/discussions/1054#discussioncomment-6016340?
If it doesn't work, you can set skip_transform
to True
in the automl_settings
and try again. It should be reproducible.
Discussed in https://github.com/microsoft/FLAML/discussions/1054
I have the same issue. I use sklearn pipeline with flaml and then reproduce with sklearn pipeline. The results are totally different. Not only rf, but also for k neighbor (without random seed effect). automl_pipeline = Pipeline([ ("standardizer", standardizer), ("automl", automl) ]) automl_settings = { "time_budget": 240, "estimator_list": ['kneighbor'], #rf "eval_method": 'cv', "split_type": 'stratified', "n_splits": 5, "metric": 'accuracy', "task": 'classification', "log_file_name": "data.log", "seed": 42, "verbose":5 }