dask-searchcv incompatible with Dask v0.18 about dask-searchcv HOT 7 CLOSED

Can you provide a reproducible example? This works for me

In [5]: from distributed import Client

In [6]: client = Client()

In [7]: from sklearn.datasets import load_digits
   ...: from sklearn.svm import SVC
   ...:
   ...: # Fit with dask-searchcv
   ...: from dask_searchcv import GridSearchCV
   ...:
   ...: param_space = {'C': [1e-4, 1, 1e4],
   ...:                'gamma': [1e-3, 1, 1e3],
   ...:                'class_weight': [None, 'balanced']}
   ...:
   ...: model = SVC(kernel='rbf')
   ...:
   ...: digits = load_digits()
   ...:
   ...: search = GridSearchCV(model, param_space, cv=3)
   ...: search.fit(digits.data, digits.target)

from dask-searchcv.

Hi @TomAugspurger . I'll try this later this evening. However, I note that you aren't providing a remote scheduler IP address in this example. Perhaps that doesn't matter, but I'll confirm.

from dask-searchcv.

from dask-searchcv.

Example of v0.17.5 working correctly:
https://www.youtube.com/watch?v=TD9elHu5ag4

Example of v0.18.1 working incorrectly:
https://www.youtube.com/watch?v=ZXxZ9NvV8C0

@TomAugspurger please let me know if I can help to provide additional information.

from dask-searchcv.

Closing this in favor of dask/dask-ml#249

from dask-searchcv.

This was fixed over in dask/dask-ml#260

It'll be included in the next release of dask-ml which is probably sometime in the next week or two.

from dask-searchcv.

Thanks

from dask-searchcv.