Hyperfrost devel

[telegraphic]

hyperfrost: combining hyperseti with bifrost. Aim is to allow basic pipelining of data reading and hit searching. Main goals:

1) A bifrost source block that reads a hyperseti DataArray 2) A hyperseti search block that connects to a bifrost pipeline. 3) A pipelined version of hyperseti where data reads are concurrent with GPU kernel execution.

WTD summary

• Added Documentation for Hyperfrost Component New documentation providing guidelines for utilizing the Hyperfrost component has been introduced in hyperfrost.md under the docs/usage directory.

• Context Management and File Closing Features Several updates have been made to manage context and ensure closure of file handlers in data_array.py, io/hdf5.py, io/hit_db.py.

• New Parameter and Feature in dedoppler.py A new parameter mm_sc and a significant functionality regarding SmearCorrMan have been incorporated in dedoppler.py.

• New Parameter in hits.py Function hitsearch() function in hits.py has been updated to include a new parameter mm.

• Additions in hyperseti Directory Five new code files have been added to the hyperseti directory associated with Hyperfrost.

• Extension of Import Functionality load_data.py function is updated to make imports of setigen only when necessary, reducing unnecessary load.

• Update of Import Statement The import statement import setigen as stg has been updated in io/setigen.py file.

• Update in kernels Directory The introduction of BlankHitsMan class has been made in kernels directory.

• Better Information Presentation on Memory/block Size The info function in kernel_manager.py has been updated to print details about memory usage and block size.

• Sorting Capability in peak_finder.py A new helper function to sort list of hits into groups has been incorporated in peak_finder.py.

• Enhanced Pipeline Functionality Introduction of optional kernel_managers into GulpPipeline class in pipeline.py improves user control on kernel managers. Methods preprocess, dedoppler, and hitsearch are updated to use kernel managers.

• Updated Package Dependencies The required version of cupy has been relaxed to >=10.0 and other package dependencies updated in pyproject.toml.

• New Test Cases for Hyperfrost Component test_bifrost.py containing test cases for Hyperfrost functionalities has been introduced.

• Expanded Test Coverage New test cases have been added to test_data_array.py and test_with_voyager.py to ensure stability and reliability of the introduced updates.

[what-the-diff[bot]]

PR Summary

• Added new kernel manager for managing data A new kernel manager was introduced to help with data processing.

• Updated dedoppler and hitsearch functions These functions now accept optional kernel manager objects for more customizability.

• Fixed duplicate logbook import bug Resolved an issue where the logbook was imported twice, causing logging problems.

• Changed default gulp size setting The default value now allows setting via config file or command line arguments.

• Added info method to kernel managers This method provides useful memory usage information for workspace variables.

• Enhanced testing with new test case An additional test case was added to improve the overall testing process.