Move thread local entry into Learner. #5396
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
Extracted from dmlc#5389 . This is an attempt to workaround CUDA context issue in static variable, where the CUDA context can be released before device vector. * Add PredictionEntry to thread local entry. This eliminates one copy of prediction vector. * Don't define CUDA C API in a namespace.
Codecov Report
@@ Coverage Diff @@
## master #5396 +/- ##
=======================================
Coverage 84.07% 84.07%
=======================================
Files 11 11
Lines 2411 2411
=======================================
Hits 2027 2027
Misses 384 384 Continue to review full report at Codecov.
|
RAMitchell
approved these changes
Mar 7, 2020
| @@ -167,6 +184,8 @@ class Learner : public Model, public Configurable, public rabit::Serializable { | |||
| virtual std::vector<std::string> DumpModel(const FeatureMap& fmap, | |||
| bool with_stats, | |||
| std::string format) const = 0; | |||
|
|
|||
| virtual XGBAPIThreadLocalEntry& GetThreadLocal() const = 0; | |||
There was a problem hiding this comment.
We could make Learner a concrete class. I don't see us subclassing different learners any time soon.
There was a problem hiding this comment.
Actually I don't like Learner, as linear and tree are so different.
| @@ -105,6 +105,17 @@ class Transform { | |||
| return Span<T const> {_vec->ConstHostPointer(), | |||
| static_cast<typename Span<T>::index_type>(_vec->Size())}; | |||
| } | |||
| // Recursive sync host | |||
There was a problem hiding this comment.
What made you do this? Just curious.
There was a problem hiding this comment.
Train hist with cupy data. I modified a test in this PR.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Extracted from #5389 .
This is an attempt to workaround CUDA context issue in static variable, where
the CUDA context can be released before device vector.
Fix training with GPU data on multi-threaded environment. Calling
HostVectorinTransformcauses race condition.Add PredictionEntry to thread local entry.
This eliminates one copy of prediction vector.