Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

mpmath creates Memory Error on import

See original GitHub issue

I wasn’t able to determine the reason, but it seems that when importing mpmath with gmp=6.2.0 it creates a Memory Error during this operation:

mpmath/mpmath/libmp/libelefun.py

# ~= next power of two + 20
cache_prec_steps = [22,22]
for k in xrange(1, bitcount(LOG_TAYLOR_PREC)+1):
cache_prec_steps += [min(2**k,LOG_TAYLOR_PREC)+20] * 2**(k-1)

When I downgraded gmp<=6.1.2.

mpmath imported successfully.

Stack trace is unavailable. If I am able to reproduce behavior I will attach.

python 3.7.6 gmpy2 2.1.0b1

Issue Analytics

State:
Created 4 years ago
Comments:15

Top GitHub Comments

1reaction

jjhelmuscommented, Jan 30, 2020

I can confirm that the 6.2.0 build 1 packages (conda-forge/linux-64::gmp-6.2.0-he1b5a44_1) give the expect results for the test program and the python test.

0reactions

brianmcconnelcommented, Feb 2, 2020

I can also confirm that this now works after updating to the latest version of gmp=6.2.0 gmp 6.2.0 he1b5a44_1 conda-forge

I was able to get it to work on two intel cpus: An old Intel® Xeon® CPU E5-2643 0 @ 3.30GHz And a newer Intel® Core™ i5-10210U CPU @ 1.60GHz

Thanks for the help in getting to the bottom of the issue.

Top Results From Across the Web

Precision and representation issues - mpmath

This section gives an overview of arbitrary-precision binary floating-point arithmetic and some concepts from numerical analysis. The following concepts are ...

Python uses huge amount of RAM to import file - Stack Overflow

I have hit a snag when I increase the matrix size, which causes the exernal file ... from __future__ import division import mpmath...

Out of memory on big expressions · Issue #17609 - GitHub

For some particular expressions, SymPy eats my 512 GB of RAM, and ends up crashing ... from sympy.parsing.sympy_parser import parse_expr ...

MemoryError when importing a custom module

When I try to import the module through the UART REPL, I get the following error: Code: Select all MemoryError: memory allocation failed, ......

Utility functions — SymPy 0.7.4.1 documentation - omz:software

from mpmath import * >>> mp.dps = 15; mp.pretty = False ... Exact addition avoids cancellation errors, enforcing familiar laws of numbers such...