Closed AlexanderLavelle closed 2 years ago
@AlexanderLavelle sounds interesting - you are right that each TabularAutoML model has the reader by default. From your code I can see that you are using the ReportDeco decorated TabularAutoML object, but this should not change the situation.
Could you please provide us with the code for the task, roles and model creation+training code? It can help us to find the key.
Alex
Task:
task = Task('binary', )
roles:
roles = {
'target': TARGET_NAME,
#'drop': ['id']
}
model:
RD = ReportDeco(output_path = 'tabularAutoML_model_report')
automl_rd = RD(
TabularAutoML(
task = task,
timeout = TIMEOUT,
cpu_limit = N_THREADS,
reader_params = {'n_jobs': N_THREADS, 'cv': N_FOLDS, 'random_state': RANDOM_STATE},
general_params={'use_algos': [['lgb', 'lgb_tuned', 'cb', 'linear_l2']]},
#lgb_params = {'default_params':{gpu_device_id':0, 'gpu_platform_id':0}},
gpu_ids=0,
memory_limit=35
)
)
prediction:
oof_pred = automl_rd.fit_predict(trainDf, roles = roles, verbose = 4)
Stale issue message
I am currently having an issue where I have replicated another project. After finishing the model, when trying to predict, I get:
I'm not sure how this error occurs as the TabularModel was instantiated with reader?
Thank you in advance