[BE] Enable Ruff's Flake8 PYI042 (#111114)

Enable [snake-case-type-alias (PYI042)](https://docs.astral.sh/ruff/rules/snake-case-type-alias/)

Link: #110950
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111114
Approved by: https://github.com/albanD

pyproject.toml

@@ -52,7 +52,6 @@ ignore = [
     "PYI034",
     "PYI036",
     "PYI041",
-    "PYI042",
     "PYI045",
     "PYI056",
     "SIM102", "SIM103", "SIM112", # flake8-simplify code styles

torch/optim/adadelta.pyi

@@ -1,9 +1,9 @@
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class Adadelta(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         rho: float = ...,
         eps: float = ...,

torch/optim/adagrad.pyi

@@ -1,9 +1,9 @@
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class Adagrad(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         lr_decay: float = ...,
         weight_decay: float = ...,

torch/optim/adam.py

@@ -2,7 +2,7 @@ from typing import List, Optional, Union, Tuple
 
 import torch
 from torch import Tensor
-from .optimizer import (Optimizer, params_t, _use_grad_for_differentiable, _get_value,
+from .optimizer import (Optimizer, ParamsT, _use_grad_for_differentiable, _get_value,
                         _stack_if_compiling, _dispatch_sqrt, _default_to_fused_or_foreach,
                         _capturable_doc, _differentiable_doc, _foreach_doc, _fused_doc,
                         _maximize_doc)
@@ -13,7 +13,7 @@ __all__ = ['Adam', 'adam']
 
 class Adam(Optimizer):
     def __init__(self,
-                 params: params_t,
+                 params: ParamsT,
                  lr: Union[float, Tensor] = 1e-3,
                  betas: Tuple[float, float] = (0.9, 0.999),
                  eps: float = 1e-8,

torch/optim/adam.pyi

@@ -2,12 +2,12 @@ from typing import Optional, Tuple, Union
 
 from torch import Tensor
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class Adam(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: Union[float, Tensor] = 1e-3,
         betas: Tuple[float, float] = (0.9, 0.999),
         eps: float = 1e-8,

torch/optim/adamax.pyi

@@ -1,11 +1,11 @@
 from typing import Tuple
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class Adamax(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         betas: Tuple[float, float] = ...,
         eps: float = ...,

torch/optim/adamw.py

@@ -2,7 +2,7 @@ import torch
 from torch import Tensor
 from .optimizer import (Optimizer, _use_grad_for_differentiable, _get_value, _dispatch_sqrt,
                         _stack_if_compiling, _capturable_doc, _differentiable_doc, _foreach_doc,
-                        _fused_doc, _maximize_doc, _default_to_fused_or_foreach, params_t)
+                        _fused_doc, _maximize_doc, _default_to_fused_or_foreach, ParamsT)
 from typing import List, Optional, Tuple, Union
 from torch.utils._foreach_utils import _get_fused_kernels_supported_devices
 
@@ -12,7 +12,7 @@ __all__ = ["AdamW", "adamw"]
 class AdamW(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: Union[float, Tensor] = 1e-3,
         betas: Tuple[float, float] = (0.9, 0.999),
         eps: float = 1e-8,

torch/optim/adamw.pyi

@@ -2,12 +2,12 @@ from typing import Optional, Tuple, Union
 
 from torch import Tensor
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class AdamW(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: Union[float, Tensor] = 1e-3,
         betas: Tuple[float, float] = (0.9, 0.999),
         eps: float = 1e-8,

torch/optim/asgd.pyi

@@ -1,9 +1,9 @@
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class ASGD(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         lambd: float = ...,
         alpha: float = ...,

torch/optim/lbfgs.pyi

@@ -1,11 +1,11 @@
 from typing import Optional
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class LBFGS(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         max_iter: int = ...,
         max_eval: Optional[int] = ...,

torch/optim/nadam.pyi

@@ -1,11 +1,11 @@
 from typing import Tuple
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class NAdam(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         betas: Tuple[float, float] = ...,
         eps: float = ...,

torch/optim/optimizer.py

@@ -204,7 +204,7 @@ def register_optimizer_step_post_hook(hook: GlobalOptimizerPostHook) -> Removabl
     _global_optimizer_post_hooks[handle.id] = hook
     return handle
 
-params_t: TypeAlias = Union[Iterable[torch.Tensor], Iterable[Dict[str, Any]]]
+ParamsT: TypeAlias = Union[Iterable[torch.Tensor], Iterable[Dict[str, Any]]]
 
 _P = ParamSpec("_P")
 R = TypeVar("R")
@@ -236,7 +236,7 @@ class Optimizer:
     _optimizer_load_state_dict_pre_hooks: 'OrderedDict[int, Callable[["Optimizer", StateDict], Optional[StateDict]]]'
     _optimizer_load_state_dict_post_hooks: 'OrderedDict[int, Callable[["Optimizer"], None]]'
 
-    def __init__(self, params: params_t, defaults: Dict[str, Any]) -> None:
+    def __init__(self, params: ParamsT, defaults: Dict[str, Any]) -> None:
         torch._C._log_api_usage_once("python.optimizer")
         self.defaults = defaults
         self._optimizer_step_pre_hooks = OrderedDict()

torch/optim/radam.pyi

@@ -1,11 +1,11 @@
 from typing import Tuple
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class RAdam(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         betas: Tuple[float, float] = ...,
         eps: float = ...,

torch/optim/rmsprop.pyi

@@ -1,9 +1,9 @@
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class RMSprop(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         alpha: float = ...,
         eps: float = ...,

torch/optim/rprop.pyi

@@ -1,11 +1,11 @@
 from typing import Tuple
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class Rprop(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         etas: Tuple[float, float] = ...,
         step_sizes: Tuple[float, float] = ...,

torch/optim/sgd.pyi

@@ -1,9 +1,9 @@
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class SGD(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float,
         momentum: float = ...,
         dampening: float = ...,

torch/optim/sparse_adam.pyi

@@ -1,11 +1,11 @@
 from typing import Tuple
 
-from .optimizer import Optimizer, params_t
+from .optimizer import Optimizer, ParamsT
 
 class SparseAdam(Optimizer):
     def __init__(
         self,
-        params: params_t,
+        params: ParamsT,
         lr: float = ...,
         betas: Tuple[float, float] = ...,
         eps: float = ...,