[IGNITE-10272] [ML] Inject learning environment into scope of dataset compute task

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Done
• Fix Version/s: 2.8
• Component/s: ml
• Assignee: Artem Malykh
• Reporter: Artem Malykh
• Created: 15/Nov/18 11:47
• Updated: 14/Jan/20 10:47
• Resolved: 04/Dec/18 12:00

Description

We want to have learing environment for each partition during training process. We need this for several reasons.

The first one is logging training process for each partition. The second reason is having the possibility of fixing random number generator seed for tests and debugging(fixing subspace distribution, stable tests, etc...)

Issue Links

• is caused by: IGNITE-8867 [ML] Bagging on learning sample
• links to: GitHub Pull Request #5484
• links to: TC build