Exception when running training with sklearn Estimator
Opened this issue · 0 comments
SamVanhoutte commented
Describe the bug
When using the estimator_type = 'sklearn'
on a gpu target, an exception is thrown:
TypeError: __init__() got an unexpected keyword argument 'use_gpu'
To Reproduce
Call the start_training
method , using sklearn
estimator_type:
args = {
'--c_value': 1.02,
'--solver': 'saga', # lbfgs newton-cg liblinear saga sag
'--train_test_split_ratio': 0.3
}
compute_run = trainer.start_training(training_name, environment_type='sklearn',
script_parameters=args, use_estimator=True,
compute_target='gpu-training', gpu_compute=True)
Expected behavior
A training that gets scheduled without error
Additional context
Add any other context about the problem here.
- Version 1.1.2.1