Fix model loading from stream #7067
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Fix bug introduced in 1791371 (allow loading from byte array) When loading model from stream, only last buffer read from the input stream is used to construct the model. This may work for models smaller than 1 MiB (if you are lucky enough to read the whole model at once), but will always fail if the model is larger.
|
Thanks for the PR! Is there a way we can add a test for this? |
Perhaps add a test that creates 1.5MB large model definition and tries to load it? I can try creating such test case, perhaps by crafting some minimal model and (as it is a json) put some generated whitespace inside to make it 1.5 MB large (and it will fail to parse if only first/last 1 MB is seen), then send it via a stream. |
|
@mpetricek-corp Better to add a unit test for your fix. Thx very much |
|
Can we generate the model with random data. To get a large model, easiest way is to use classification with random forest. |
|
Close #7168 . |
|
Merged since it's difficult to create unittest and is confirmed to work. |
NvTimLiu
pushed a commit
to NvTimLiu/spark-xgboost
that referenced
this pull request
Nov 1, 2021
Fix bug introduced in 1791371 (allow loading from byte array) When loading model from stream, only last buffer read from the input stream is used to construct the model. This may work for models smaller than 1 MiB (if you are lucky enough to read the whole model at once), but will always fail if the model is larger.
NvTimLiu
added a commit
to NvTimLiu/spark-xgboost
that referenced
this pull request
Nov 1, 2021
Fix model loading from stream (dmlc#7067) See merge request nvspark/xgboost!391
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Fix bug introduced in 1791371 (allow loading from byte array)
When loading model from stream, only last buffer read from the input stream is used to construct the model.
This may work for models smaller than 1 MiB (if you are lucky enough to read the whole model at once), but will always fail if the model is larger.