TPC NN Clusterization: CCDB support + cosmetic changes #14069
Closed
Conversation
This file contains hidden or 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
Contributor
|
REQUEST FOR PRODUCTION RELEASES: This will add The following labels are available |
Please consider the following formatting changes to AliceO2Group#14069
…ce and one initialization
ChSonnabend
changed the title
Please consider the following formatting changes to AliceO2Group#14069
ChSonnabend
added a commit
to ChSonnabend/AliceO2
that referenced
this pull request
Mar 27, 2025
ChSonnabend
added a commit
to ChSonnabend/AliceO2
that referenced
this pull request
Mar 29, 2025
… will merge AliceO2Group#14069 to have the changes in GPUChainTrackingClusterizer.
davidrohr
pushed a commit
that referenced
this pull request
Apr 20, 2025
* Initial set of bug.fixes and cosmetic changes * Please consider the following formatting changes * Adjusting eval sizes. Makes code neater and avoids some calculations * Adding separate functions. Now the host process only needs one instance and one initialization * First version of CCDB implementation * Working CCDB API calls (tested with test-ccdb) * Improve fetching, but have to pass settings by value, not const ref * Using const ref and moving CCDB calls to host initialization * Simplifications and renaming * Please consider the following formatting changes * First version of GPU stream implementation. Still needs testing. * Fixes * Please consider the following formatting changes * Adding the lane variable. This PR will in any case conflict with #14069 * Compiles on EPNs. Need to add shadow processors next. But for this, I will merge #14069 to have the changes in GPUChainTrackingClusterizer. * Adding shadow instance. Not sure if this correctly allocates GPU memory using AllocateRegisteredMemory * This runs, but will eventually fill up the VRAM. Need to include a mem clean * Found the stream allocation issue. Now starting optimizations * Improve readability and adapt for some comments * Fixing memory assignment issue. Reconstruction runs through with FP32 networks * Major reworkings to add FP16 support * Bug-fixes * Improved data filling speeds by factor 3 * Limiting threads for ONNX evaluation * Bug-fix for correct thread assignment and input data filling * Minor changes * Adding I** inference, potentally needed for CNN + FC inference * CCDB fetching of NNs ported to GPUWorkflowSpec * Adjusting CPU threads and ORT copmile definitions * About 10x speed-up due to explicit io binding * Changes for synchronization and consistency. No performance loss. * Please consider the following formatting changes * Fixing warnings (errors due to size_t) * Fixing linker issues * Adding volatile memory allocation and MockedOrtAllocator. Removing print statements and time measurements * Please consider the following formatting changes * Circumvent "unused result" warning and build failure * Adjust for comments * Please consider the following formatting changes * Fixing build flags --------- Co-authored-by: ALICE Action Bot <alibuild@cern.ch>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
No description provided.