CODEWITHSUNDEEP
dasklightgbmdask-dataframecudf
Bala.vrad
Bala.vrad

Reputation: 53

How do I train a DaskLGBMClassifier using dask-cudf dataframe

I have a dask_cudf dataframe on which I have applied train test split to obtain the respective train, test data.

I am using DaskLGBMClassifier to train a classifier model & I can use dask dataframe to fit this model without issues

model.fit(X_train_dask.values, y_train_dask.values)

However error occurs when I do this

model.fit(X_train_dask_cudf.values, y_train_dask_cudf.values)

Error is


You have supplied a custom function and Dask is unable to 
determine the type of output that that function returns. 

To resolve this please provide a meta= keyword.
The docstring of the Dask function you ran should have more information.

Original error is below:
------------------------
RuntimeError('CuPy failed to load libnvrtc.so.12: OSError: libnvrtc.so.12: cannot open shared object file: No such file or directory')

Also tried fitting using the dask_cudf dataframe without converting to arrays. But got below error

TypeError: Data must be one of: numpy arrays, pandas dataframes, sparse matrices (from scipy). Got DataFrame.

Does someone know what could be going wrong here ?.

Upvotes: 0

Views: 87

Answers (0)

Related Questions