- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
02-10-2021 05:53 AM
Hello all,
I have created a new WORD CORPUS and a new Solution Definition for Clustering. I SUBMIT and TRAIN my solution definition and it returned back with a training error.
Failure Message: SE0060:Solution Training Failed. Ask Support to use log key 463366 to investigate trainer logs further.
What should I look for ?
Solved! Go to Solution.
- Labels:
-
Predictive Intelligence

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
02-10-2021 06:23 AM
Is this a Personal Developer Instance(PDI) or a sub-prod instance? If it's sub-prod open a Hi ticket and include the log key error. Support has access to look up the log for your PI job on the PI training server - customers don't have access to that log. You can ask them to dump the log into the ticket so you can see what they are looking at.
If this is a PDI, we don't support training the solution definitions on PDIs yet.
-Lener
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-26-2021 04:43 AM
Dear all,
I have the same problem and would be keen to get a solution.
Highly appreciated.
Martin
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-26-2022 06:35 AM
Good Afternoon,
I am having a similar problem but with a different error log:
SE0060:Solution Training Failed. Ask Support to use log key 352893 to investigate trainer logs further.
Do you remember how you got this resolved?
Kind Regards
Ashley

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-26-2022 06:50 AM
While waiting for support to review you can check a few things:
1. Make sure the data in your word corpus & training set has valid data. You should make sure nulls are not in the data or special characters as that will cause an issue when training. Your data should also be of the support types for PI, such as string, reference, choice.
2. Make sure your word corpus and training set is within the default max limits, which is 300k rows. Make sure you have enough data in the cluster, default min is 10k.
3. Also I am assuming that you are not changing any of the default settings. There are advanced settings but that should be used after you have completed a successful training.