Skip to content

numpy/argument-in-wrong-slot

Rule argument-in-wrong-slot
Pack NumPy
Severity error
Fix unsafe, with xpd check --fix --unsafe-fixes
Library versions every version

These calls put an argument where NumPy expects another one, such as a second size where np.zeros expects the dtype, so they raise or give a wrong result. Pass the shape as one tuple, or pass the argument in its own place.

grid = np.zeros(3, 4)
grid = np.zeros((3, 4))

The rule reports np.zeros(3, 4), np.full(3, 4, 0), np.reshape(a, 2, 3), np.vstack(a, b), np.concatenate(a, [4, 5]), np.random.rand((3, 4)), np.random.random(3, 4), np.linspace(0, 1, 0.1) and np.quantile(x, 95). Each call puts an argument in the place of another parameter, so it raises.

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 writes np.zeros((3, 4)), np.hstack((a, b, a)) with any keywords after the tuple, and np.percentile with the same arguments for a quantile call with more arguments. The linspace form has no fix, because the code does not show whether a count or a step was meant.

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

The second argument of $NP.$F is the dtype, so $NP.$F($A, $B) raises TypeError. Pass the shape as one tuple, $NP.$F(($A, $B)).

Browse the other rules of the NumPy pack.