Torsten Wörtwein
Torsten Wörtwein
**Feature** When the implementation adds new arguments (or the stubs forget some of the arguments), stubtest creates many errors if the new/forgotten argument isn't the last argument. ```py $ cat...
**To Reproduce** ```py from typing import Literal def fun(**kwargs) -> None: ... class A: def __init__(self): self.a: Literal["a"] = "a" def test(self) -> None: return fun(**{self.a: "a"}) # error: Keywords...
## 🐛 Bug TorchScript wants to derive a name from a function using `__name__`. `functools.partial` does not have `__name__` but has `name`. ## To Reproduce ```py from functools import partial...
The following does not work with 3.11.1 at runtime: test.py ```py from typing import TYPE_CHECKING if TYPE_CHECKING: from test_b import B def test(x: B) -> B: return x ``` test_b.py...
Folders containing only type stubs should not need `__init__.py` (not sure: but they might need an `__init__.pyi` )
`FaceLandmarkers` is two times slower than `FaceMesh` even though both return only the facial landmarks in their default configuration. ```py import mediapipe as mp from mediapipe.python.solutions.face_mesh import FaceMesh from mediapipe.tasks...
### Description of issue (what needs changing) Document performance differences between facial landmark models. ### Clear description `FaceMesh` return landmarks, which can optionally be refined (`refine_landmarks`). It is not clear...
## 🐞Describing the bug I want to train a model that uses MFCCs as the input. Luckily CoreMLTools is able to convert torchaudio's MFCC :) but there are some numerical...
I think a few libraries that are checked by mypy-primer use pandas. It might be nice to install pandas-stubs as a dependency for those projects. Background: It might be nice...