ludwig ray datasource checkin#4
Conversation
ludwig/backend/ray.py
Outdated
|
|
||
|
|
||
| # Load model from checkpoint | ||
| # Load model from checkpoint |
ludwig/backend/ray.py
Outdated
| logger.info("###### checkpoint: {}".format(trainer_results.checkpoint)) # ***FIXME | ||
| print("###### checkpoint: {}".format(trainer_results.checkpoint)) # ***FIXME |
ludwig/backend/ray.py
Outdated
| if "validation_field" in trainer_results.metrics: | ||
| self._validation_field = trainer_results.metrics["validation_field"] | ||
| else: | ||
| self._validation_field = None | ||
|
|
||
| if "validation_metric" in trainer_results.metrics: | ||
| self._validation_metric = trainer_results.metrics["validation_metric"] | ||
| else: | ||
| self._validation_metric = None | ||
|
|
There was a problem hiding this comment.
I got a Key error exception while testing one our solutions
There was a problem hiding this comment.
which solution?
what is the exact error that it is complaining about? can you paste the stack trace.
can you fix it by changing the ML job file parameters (for example. see git issue 2458)
There was a problem hiding this comment.
I will change the parameters in the ml job file, reverted the code
Unit Test Results 6 files ±0 6 suites ±0 45m 54s ⏱️ - 28m 57s For more details on these failures, see this check. Results for commit 67db94f. ± Comparison against base commit 4491d65. ♻️ This comment has been updated with latest results. |
Code Pull Requests
Please provide the following: