cf.Data.fits_in_memory

Data.fits_in_memory()[source]

Return True if the array is small enough to be retained in memory.

Returns True if the size of the computed array, always including space for a full boolean mask, is small enough to be retained in available memory.

Performance

The delayed operations are actually not computed by fits_in_memory, so it is possible that an intermediate operation may require more than the available memory, even if the final array does not.

See also

array, compute, nbytes, persist, cf.free_memory

Parameters
itemsize: deprecated at version 3.14.0

The number of bytes per word of the master data array.

Returns
bool

Whether or not the computed array fits in memory.

Examples

>>> d = cf.Data([1], 'm')
>>> d.fits_in_memory()
True

Create a double precision (8 bytes per word) array that is approximately twice the size of the available memory:

>>> size = int(2 * cf.free_memory() / 8)
>>> d = cf.Data.empty((size,), dtype=float)
>>> d.fits_in_memory()
False
>>> d.nbytes * (1 + 1/8) > cf.free_memory()
True