[MLEARN-3] Store ranking neural network labels in non-flat way in JCR Created: 10/Dec/18 Updated: 06/Feb/19 Resolved: 06/Feb/19 |
|
| Status: | Closed |
| Project: | Machine Learning |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Neutral |
| Reporter: | Cedric Reichenbach | Assignee: | Unassigned |
| Resolution: | Outdated | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Template: |
|
||||||||
| Acceptance criteria: |
Empty
|
||||||||
| Task DoR: |
Empty
|
||||||||
| Date of First Response: | |||||||||
| Epic Link: | Periscope improvements | ||||||||
| Description |
|
Storing 1000+ nodes or properties on the same level leads to performance issues with JCR/Jackrabbit. Since we potentially store more labels than that (up to 10k by default), we should do so in a hierarchical structure with always less than 1k neighbours on any level. Current location of labels (on demo): https://demo.magnolia-cms.com/.magnolia/admincentral#app:configuration:browser;/modules/periscope-result-ranker/persistence/labels:treeview: |
| Comments |
| Comment by Mikaël Geljić [ 08/Jan/19 ] |
|
quoting relevant input from the Slack thread over #pd-foundation:
|