Skip to content

numpy/einsum-optimize

Rule einsum-optimize
Pack NumPy
Severity warning
Fix unsafe, with xpd check --fix --unsafe-fixes
Library versions every version

Without optimize=, einsum loops over every index at once, which was 339 times slower for three 100 by 100 matrices. Add optimize=True.

chain = np.einsum("ij,jk,kl->il", a, b, c)
chain = np.einsum("ij,jk,kl->il", a, b, c, optimize=True)

The rule reports np.einsum with three or more 2-D operands and a 2-D result, such as "ij,jk,kl->il", and no optimize=.

xpd check --fix --unsafe-fixes applies the fixes of this rule. Check the result, because an unsafe fix can change what the code does.

The unsafe fix adds optimize=True.

xpd prints this message with the issue. A name that starts with $ stands for code from your file.

Without optimize= einsum loops over every index at once, which ran 339 times slower on three 100 by 100 matrices. Add optimize=True.

Browse the other rules of the NumPy pack.