Open cjnolet opened 4 years ago
This issue has been marked rotten due to no recent activity in the past 90d. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.
This issue has been marked stale due to no recent activity in the past 30d. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be marked rotten if there is no activity in the next 60d.
Random projection currently stores internal parameters as an internal object that is created in the cython layer but opaque to the python layer. This means pickling the random projection model will not actually pickle the trainer parameters.
To follow the design of the rest of the cuml models, random projection should instead of CumlArray to manage the arrays that comprise its trained state so that it can be properly pickled.