ashdisperse.config package

Submodules

ashdisperse.config.config module

ashdisperse.config.config.get_chunk_size() → int

Get the current chunk size used for processing.

This function retrieves the value of the internal variable _chunk_size, which determines the number of items processed in each chunk.

Returns:

The chunk size.

Return type:

int

Example:

>>> ashdisperse.config.get_chunk_size()
>>> 8

ashdisperse.config.config.get_max_threads() → int

Get the maximum number of threads available on the system

Returns:

maximum number of threads

Return type:

int

Example:

>>> ashdisperse.config.get_max_threads()
>>> 8

ashdisperse.config.config.get_num_threads() → int

Get the number of threads available for computation.

This function retrieves the number of threads that can be used, typically for parallel processing, by calling the underlying numba.get_num_threads() function.

Returns:

The number of available threads for parallel processing.

Return type:

int

Example:

>>> ashdisperse.config.get_num_threads()
>>> 4

ashdisperse.config.config.get_threading_layer() → str

Get the numba threading layer

Returns:

threading layer used by numba

Return type:

str

Example:

>>> ashdisperse.config.get_theading_layer()
>>> "omp"

ashdisperse.config.config.set_chunk_size(n: int = 8) → None

Set the global chunk size used for processing.

Parameters:

n (int, optional) – The desired chunk size. Must be a positive integer. Defaults to 8.

Example:

>>> ashdisperse.config.set_chunk_size(16)  # Sets the chunk size to 16
>>> ashdisperse.config.get_chunk_size()
>>> 16

ashdisperse.config.config.set_default_threads(n=None)

Set the default number of threads for processing.

Parameters:

n (int or None, optional) – Number of threads to set. If None, sets to one less than the maximum available threads.

Example:

>>> ashdisperse.config.set_default_threads()      # Sets threads to max_threads - 1
>>> ashdisperse.config.get_default_threads()
>>> 7

>>> ashdisperse.config.set_default_threads(4)     # Sets threads to 4
>>> ashdisperse.config.get_default_threads()
>>> 4

ashdisperse.config.config.set_num_threads(n)

Set the number of threads to be used for computation, ensuring it does not exceed the maximum recommended threads. Issues warnings if the requested number of threads is greater than or equal to the maximum, and adjusts accordingly.

Also attempts to set MKL threads locally to 1 for compatibility.

Parameters:

n (int) – The desired number of threads to use for computation.

Example:

>>> ashdisperse.config.set_num_threads(6)
>>> ashdisperse.config.get_num_threads()
>>> 6

ashdisperse.config.config.set_threading_layer(thread_layer: str = 'tbb') → None

Set the numba threading layer

Parameters:

thread_layer (str, optional) – threading layer selected from “tbb”, “omp” and “workqueue”, defaults to “tbb”

Raises:

ValueError – if thread_layer not one of “tbb”, “omp” or “workqueue”

Example:

>>> ashdisperse.config.set_threading_layer("omp")
>>> ashdisperse.config.get_threading_layer()
>>> "omp"

Module contents