Load initial scores with binary data files in CLI version #4807
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is to fix #4786. Currently there's no way to load initial score when loading from binary data file in CLI version.
In
Metadata::LoadFromMemory
andMetadata::SaveBinaryToFile
, other fields ofMetadata
is handled, exceptinit_score_
. The only way to set initial score when loading from binary data files is the call methods likeset_init_score
in Python, which is impossible for CLI version.The implementation goes like this: after loading from binary data files, checks whether there are initial score files exists. If so, initial scores are loaded.
The corresponding documentation for continual training is also updated.