Skip to content

Instantly share code, notes, and snippets.

@maggiemoss

maggiemoss/pyrefly_before_2n

Created October 2, 2025 22:06
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save maggiemoss/911b4d0bc88bf8cf3ab91f67184e9d46 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save maggiemoss/911b4d0bc88bf8cf3ab91f67184e9d46 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
pyrefly_before_2n
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:115:28
|
115 | def __init__(self, op: RedOpType) -> None: ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:117:10
|
117 | SUM: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:118:10
|
118 | AVG: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:119:14
|
119 | PRODUCT: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:120:10
|
120 | MIN: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:121:10
|
121 | MAX: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:122:11
|
122 | BAND: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:123:10
|
123 | BOR: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:124:11
|
124 | BXOR: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:125:17
|
125 | PREMUL_SUM: RedOpType = ...
| ^^^^^^^^^
|
ERROR Could not find name `RedOpType` [unknown-name]
--> torch/_C/_distributed_c10d.pyi:126:13
|
126 | UNUSED: RedOpType = ...
| ^^^^^^^^^
|
ERROR `OpOverload[Ellipsis, Any] | OpOverloadPacket[Ellipsis, Any] | OperatorBase` is not assignable to `OperatorBase` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_decomp/__init__.py:243:5
|
243 | / for opo in registry:
244 | | if isinstance(opo, (OpOverload, OpOverloadPacket)):
245 | | packets_to_overloads[opo.overloadpacket].append(opo)
| |________________________________________________________________^
|
ERROR `**` is not supported between `Tensor` and `Literal[2]` [unsupported-operation]
--> torch/_decomp/decompositions.py:385:12
|
385 | loss = (self - target) ** 2
| ^^^^^^^^^^^^^^^^^^^^
|
Argument `Literal[2]` is not assignable to parameter with type `TensorBase`
ERROR `**` is not supported between `Tensor` and `Literal[2]` [unsupported-operation]
--> torch/_decomp/decompositions.py:385:12
|
385 | loss = (self - target) ** 2
| ^^^^^^^^^^^^^^^^^^^^
|
Expected 1 more positional argument
ERROR `**` is not supported between `Tensor` and `Literal[2]` [unsupported-operation]
--> torch/_decomp/decompositions.py:418:43
|
418 | loss = torch.where(loss < beta, 0.5 * loss**2 / beta, loss - 0.5 * beta)
| ^^^^^^^
|
Argument `Literal[2]` is not assignable to parameter with type `TensorBase`
ERROR `**` is not supported between `Tensor` and `Literal[2]` [unsupported-operation]
--> torch/_decomp/decompositions.py:418:43
|
418 | loss = torch.where(loss < beta, 0.5 * loss**2 / beta, loss - 0.5 * beta)
| ^^^^^^^
|
Expected 1 more positional argument
ERROR Argument `Unknown | None` is not assignable to parameter `dim` with type `Size | int | list[int] | tuple[int, ...]` in function `torch._C._VariableFunctions.amin` [bad-argument-type]
--> torch/_decomp/decompositions.py:4896:33
|
4896 | amin = torch.amin(self, dim=dim, keepdim=keepdim)
| ^^^
|
ERROR Argument `Unknown | None` is not assignable to parameter `dim` with type `Size | int | list[int] | tuple[int, ...]` in function `torch._C._VariableFunctions.amax` [bad-argument-type]
--> torch/_decomp/decompositions.py:4897:33
|
4897 | amax = torch.amax(self, dim=dim, keepdim=keepdim)
| ^^^
|
ERROR `*` is not supported between `Tensor` and `Number` [unsupported-operation]
--> torch/_decomp/decompositions.py:5141:18
|
5141 | result = result * alpha
| ^^^^^^^^^^^^^^
|
Argument `Number` is not assignable to parameter `other` with type `Tensor | bool | complex | float | int` in function `torch._C.TensorBase.__mul__`
ERROR No matching overload found for function `os.fspath` [no-matching-overload]
--> torch/_library/fake_profile.py:248:22
|
248 | f = os.fspath(f)
| ^^^
|
Possible overloads:
(path: str) -> str [closest match]
(path: bytes) -> bytes
(path: PathLike[AnyStr]) -> AnyStr
ERROR No matching overload found for function `os.fspath` [no-matching-overload]
--> torch/_library/fake_profile.py:312:22
|
312 | f = os.fspath(f)
| ^^^
|
Possible overloads:
(path: str) -> str [closest match]
(path: bytes) -> bytes
(path: PathLike[AnyStr]) -> AnyStr
ERROR Argument `str` is not assignable to parameter `object` with type `LiteralString` in function `list.append` [bad-argument-type]
--> torch/_library/infer_schema.py:162:27
|
162 | params.append(f"{schema_type} {name}")
| ^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Argument `str` is not assignable to parameter `object` with type `LiteralString` in function `list.append` [bad-argument-type]
--> torch/_library/infer_schema.py:179:27
|
179 | params.append(f"{schema_type} {name}={default_repr}")
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Expected a type form, got instance of `_SpecialForm | type` [not-a-type]
--> torch/_library/infer_schema.py:205:26
|
205 | (typing.Optional[base_type], f"{cpp_type}?"),
| ^^^^^^^^^
|
ERROR Expected a type form, got instance of `_SpecialForm | type` [not-a-type]
--> torch/_library/infer_schema.py:223:62
|
223 | for seq_typ in derived_seq_types(typing.Optional[base_type])
| ^^^^^^^^^
|
ERROR Cannot index into `dict[type[Tensor] | type[bool] | type[float] | type[int] | type[list[Tensor]] | TypeAlias[Number, type[bool | float | int]], str]` [index-error]
--> torch/_library/infer_schema.py:276:39
|
276 | return SUPPORTED_RETURN_TYPES[annotation]
| ^^^^^^^^^^
|
Argument `type[_empty] | Unknown` is not assignable to parameter `key` with type `type[Tensor] | type[bool] | type[float] | type[int] | type[list[Tensor]] | TypeAlias[Number, type[bool | float | int]]` in function `dict.__getitem__`
ERROR `list[SymInt | int]` is not assignable to variable `stride` with type `int | list[int]` [bad-assignment]
--> torch/_meta_registrations.py:2344:18
|
2344 | stride = [stride] * len(dims)
| ^^^^^^^^^^^^^^^^^^^^
|
ERROR `list[SymInt | int]` is not assignable to variable `padding` with type `int | list[int]` [bad-assignment]
--> torch/_meta_registrations.py:2349:19
|
2349 | padding = [padding] * len(dims)
| ^^^^^^^^^^^^^^^^^^^^^
|
ERROR `list[SymInt | int]` is not assignable to variable `dilation` with type `int | list[int]` [bad-assignment]
--> torch/_meta_registrations.py:2354:20
|
2354 | dilation = [dilation] * len(dims)
| ^^^^^^^^^^^^^^^^^^^^^^
|
ERROR `list[SymInt | int]` is not assignable to variable `output_padding_list` with type `list[int] | None` [bad-assignment]
--> torch/_meta_registrations.py:2361:35
|
2361 | output_padding_list = [output_padding] * len(dims)
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2373:21
|
2373 | padding[i],
| ^^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2374:21
|
2374 | dilation[i],
| ^^^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2376:21
|
2376 | stride[i],
| ^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2382:35
|
2382 | _formula(dims[i], padding[i], dilation[i], kernel_size[i], stride[i])
| ^^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2382:47
|
2382 | _formula(dims[i], padding[i], dilation[i], kernel_size[i], stride[i])
| ^^^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot index into `int` [index-error]
--> torch/_meta_registrations.py:2382:76
|
2382 | _formula(dims[i], padding[i], dilation[i], kernel_size[i], stride[i])
| ^^^^^^^^^
|
Object of class `int` has no attribute `__getitem__`
ERROR Cannot set item in `list[list[int]]` [unsupported-operation]
--> torch/_meta_registrations.py:3447:9
|
3447 | strides[len(before_shape) : len(self.shape) - len(after_shape)] = [0] * len(
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
No matching overload found for function `list.__setitem__`
Possible overloads:
(key: SupportsIndex, value: list[int], /) -> None [closest match]
(key: slice[Any, Any, Any], value: Iterable[list[int]], /) -> None
ERROR Object of class `NoneType` has no attribute `device` [missing-attribute]
--> torch/_meta_registrations.py:6658:13
|
6658 | x.device == input_gates.device
| ^^^^^^^^
|
ERROR `device | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims/__init__.py:430:5
|
430 | / for arg in args_:
431 | | if isinstance(arg, TensorLike):
432 | | if utils.is_cpu_scalar_tensor(arg):
433 | | if device is None:
434 | | device = arg.device
435 | | # keep going, in case there is a cuda tensor later
| |___________________________________________________________________^
|
ERROR Argument `SymInt | Tensor | bool | int | Unknown` is not assignable to parameter `input` with type `Tensor | bool | float | int` in function `torch._C._VariableFunctions.div` [bad-argument-type]
--> torch/_prims/__init__.py:1018:26
|
1018 | return torch.div(a, b, rounding_mode="trunc")
| ^
|
ERROR Argument `SymInt | Tensor | bool | int | Unknown` is not assignable to parameter `input` with type `Tensor | bool | float | int` in function `torch._C._VariableFunctions.true_divide` [bad-argument-type]
--> torch/_prims/__init__.py:1020:34
|
1020 | return torch.true_divide(a, b)
| ^
|
ERROR Expected *-unpacked _P.args and **-unpacked _P.kwargs [invalid-param-spec]
--> torch/_prims/context.py:128:33
|
128 | return orig_func(*args, **kwargs)
| ^^^^^^^^^^^^^^^^^
|
ERROR Expected *-unpacked _P.args and **-unpacked _P.kwargs [invalid-param-spec]
--> torch/_prims/context.py:150:33
|
150 | return orig_func(*args, **kwargs)
| ^^^^^^^^^^^^^^^^^
|
ERROR Expected *-unpacked _P.args and **-unpacked _P.kwargs [invalid-param-spec]
--> torch/_prims/context.py:158:25
|
158 | return orig_func(*args, **kwargs)
| ^^^^^^^^^^^^^^^^^
|
ERROR Object of class `FunctionType` has no attribute `register_fake` [missing-attribute]
--> torch/_prims/rng_prims.py:32:5
|
32 | rngprim_def.register_fake(impl_meta)
| ^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `device` [missing-attribute]
--> torch/_prims/rng_prims.py:332:22
|
332 | device_idx = rng_state.device.index
| ^^^^^^^^^^^^^^^^
|
ERROR Argument `Unknown | None` is not assignable to parameter `_new_state` with type `Generator` in function `torch._C.Generator.graphsafe_set_state` [bad-argument-type]
--> torch/_prims/rng_prims.py:335:39
|
335 | generator.graphsafe_set_state(rng_state)
| ^^^^^^^^^
|
ERROR `SymInt | int` is not assignable to `int` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:116:5
|
116 | / for x, y in zip(a, b):
117 | | if allow_rhs_unbacked:
118 | | if isinstance(y, torch.SymInt):
119 | | continue
120 | |
121 | | # if we do not know, then they are not the same.
| |_________________________________________________________^
|
ERROR Function declared to return `bool`, but one or more paths are missing an explicit `return` [bad-return]
--> torch/_prims_common/__init__.py:393:6
|
393 | ) -> bool:
| ^^^^
|
ERROR Function declared to return `tuple[int, ...]` but is missing an explicit `return` [bad-return]
--> torch/_prims_common/__init__.py:814:6
|
814 | ) -> tuple[int, ...]:
| ^^^^^^^^^^^^^^^
|
ERROR Function declared to return `int` but is missing an explicit `return` [bad-return]
--> torch/_prims_common/__init__.py:819:77
|
819 | def canonicalize_dims(rank: int, indices: int, wrap_scalar: bool = True) -> int:
| ^^^
|
ERROR `device | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:865:5
|
865 | / for arg in args:
866 | | if isinstance(arg, Number):
867 | | continue
868 | | elif isinstance(arg, TensorLike):
869 | | if allow_cpu_scalar_tensors and is_cpu_scalar_tensor(arg):
870 | | continue
| |_________________________^
|
ERROR `Size | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:912:5
|
912 | / for arg in args:
913 | | if isinstance(arg, Number):
914 | | continue
915 | | elif isinstance(arg, TensorLike):
916 | | if allow_cpu_scalar_tensors and is_cpu_scalar_tensor(arg):
917 | | continue
| |_________________________^
|
ERROR `Size | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:938:5
|
938 | / for arg in args:
939 | | if isinstance(arg, Number):
940 | | continue
941 | | elif isinstance(arg, TensorLike):
942 | | if allow_cpu_scalar_tensors and is_cpu_scalar_tensor(arg):
943 | | scalar_shape = arg.shape
| |_________________________________________^
|
ERROR `Sequence[int] | Size | list[int] | tuple[int, ...]` is not assignable to variable `shape` with type `Size | list[int] | tuple[Size | list[int] | tuple[int, ...]] | tuple[int, ...]` [bad-assignment]
--> torch/_prims_common/__init__.py:994:17
|
994 | shape = shape[0]
| ^^^^^^^^
|
ERROR Argument `SymBool | SymFloat | SymInt | Tensor | bool | complex | dtype | float | int | None` is not assignable to parameter `x` with type `Tensor | bool | complex | dtype | float | int | None` in function `_extract_dtype` [bad-argument-type]
--> torch/_prims_common/__init__.py:1295:27
|
1295 | a, b = _extract_dtype(a), _extract_dtype(b)
| ^
|
ERROR Argument `SymBool | SymFloat | SymInt | Tensor | bool | complex | dtype | float | int | None` is not assignable to parameter `x` with type `Tensor | bool | complex | dtype | float | int | None` in function `_extract_dtype` [bad-argument-type]
--> torch/_prims_common/__init__.py:1295:46
|
1295 | a, b = _extract_dtype(a), _extract_dtype(b)
| ^
|
ERROR `dtype | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:1390:5
|
1390 | / for arg in args:
1391 | | if isinstance(arg, Number):
1392 | | # Scalar type checking is disabled (and may be removed in the future)
1393 | | continue
1394 | | # if scalar_type is None:
1395 | | # scalar_type = type(arg)
| |__________________________________________^
|
ERROR `type | None` is not assignable to `None` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/_prims_common/__init__.py:1390:5
|
1390 | / for arg in args:
1391 | | if isinstance(arg, Number):
1392 | | # Scalar type checking is disabled (and may be removed in the future)
1393 | | continue
1394 | | # if scalar_type is None:
1395 | | # scalar_type = type(arg)
| |__________________________________________^
|
ERROR Returned type `Tensor | bool | complex | dtype | float | int` is not assignable to declared return type `dtype | None` [bad-return]
--> torch/_prims_common/__init__.py:1660:20
|
1660 | return one_plus_dim_tensor_dtype
| ^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Returned type `Tensor | bool | complex | dtype | float | int | None` is not assignable to declared return type `dtype | None` [bad-return]
--> torch/_prims_common/__init__.py:1662:16
|
1662 | return zero_dim_tensor_dtype
| ^^^^^^^^^^^^^^^^^^^^^
|
ERROR Function declared to return `Tensor` but is missing an explicit `return` [bad-return]
--> torch/_prims_common/wrappers.py:31:71
|
31 | def _maybe_convert_to_dtype(a: TensorLikeType, dtype: torch.dtype) -> TensorLikeType:
| ^^^^^^^^^^^^^^
|
ERROR Function declared to return `bool | complex | float | int` but is missing an explicit `return` [bad-return]
--> torch/_prims_common/wrappers.py:36:67
|
36 | def _maybe_convert_to_dtype(a: NumberType, dtype: torch.dtype) -> NumberType:
| ^^^^^^^^^^
|
ERROR Function declared to return `Sequence[Unknown]` but is missing an explicit `return` [bad-return]
--> torch/_prims_common/wrappers.py:41:65
|
41 | def _maybe_convert_to_dtype(a: Sequence, dtype: torch.dtype) -> Sequence:
| ^^^^^^^^
|
ERROR Expected 1 positional argument, got 2 in function `tuple.__new__` [bad-argument-count]
--> torch/_prims_common/wrappers.py:279:48
|
279 | f"return_types_{fn.__name__}", [(o, TensorLikeType) for o in out_names]
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR `Args[_P]` is not subscriptable [unsupported-operation]
--> torch/_prims_common/wrappers.py:297:63
|
297 | if isinstance(out, TensorLike) and isinstance(args[0], TensorLike):
| ^^^^^^^
|
ERROR Class member `BackwardsNotSupported.forward` overrides parent class `Function` in an inconsistent manner [bad-override]
--> torch/_prims_common/wrappers.py:432:13
|
432 | def forward(ctx, args_spec, *flat_args):
| ^^^^^^^
|
`BackwardsNotSupported.forward` has type `(ctx: Unknown, args_spec: Unknown, *flat_args: Unknown) -> Unknown`, which is not consistent with `(*args: Any, **kwargs: Any) -> Any` in `Function.forward` (the type of read-write attributes cannot be changed)
ERROR Expected *-unpacked _P.args and **-unpacked _P.kwargs [invalid-param-spec]
--> torch/_prims_common/wrappers.py:480:24
|
480 | result = fn(*args_, **kwargs)
| ^^^^^^^^^^^^^^^^^^
|
ERROR Expected *-unpacked _P.args and **-unpacked _P.kwargs [invalid-param-spec]
--> torch/_prims_common/wrappers.py:484:18
|
484 | return fn(*args, **kwargs)
| ^^^^^^^^^^^^^^^^^
|
ERROR Returned type `_Wrapped[_P, _T, [*args: Unknown, **kwargs: Unknown], Unknown]` is not assignable to declared return type `(ParamSpec(_P)) -> bool | complex | float | int | _T` [bad-return]
--> torch/_prims_common/wrappers.py:487:12
|
487 | return _fn
| ^^^
|
ERROR Argument `int | Any | None` is not assignable to parameter `op_type` with type `int` in function `Operand.__new__` [bad-argument-type]
--> torch/backends/_nnapi/serializer.py:417:21
|
417 | op_type=op_type,
| ^^^^^^^
|
ERROR `None` is not subscriptable [unsupported-operation]
--> torch/backends/_nnapi/serializer.py:1737:61
|
1737 | self.compute_operand_shape(out_id, dim, size_arg[dim - 2])
| ^^^^^^^^^^^^^^^^^
|
ERROR `None` is not subscriptable [unsupported-operation]
--> torch/backends/_nnapi/serializer.py:1742:32
|
1742 | f"int({scale_arg[dim - 2]} * {flex_name(image_id, dim)})",
| ^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `getVersionInt` [missing-attribute]
--> torch/backends/cudnn/__init__.py:37:31
|
37 | __cudnn_version = _cudnn.getVersionInt()
| ^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `getRuntimeVersion` [missing-attribute]
--> torch/backends/cudnn/__init__.py:38:31
|
38 | runtime_version = _cudnn.getRuntimeVersion()
| ^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `getCompileVersion` [missing-attribute]
--> torch/backends/cudnn/__init__.py:39:31
|
39 | compile_version = _cudnn.getCompileVersion()
| ^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `is_cuda` [missing-attribute]
--> torch/backends/cudnn/__init__.py:47:43
|
47 | elif runtime_major < 7 or not _cudnn.is_cuda:
| ^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `RNNMode` [missing-attribute]
--> torch/backends/cudnn/rnn.py:15:20
|
15 | return int(_cudnn.RNNMode.rnn_relu)
| ^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `RNNMode` [missing-attribute]
--> torch/backends/cudnn/rnn.py:17:20
|
17 | return int(_cudnn.RNNMode.rnn_tanh)
| ^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `RNNMode` [missing-attribute]
--> torch/backends/cudnn/rnn.py:19:20
|
19 | return int(_cudnn.RNNMode.lstm)
| ^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `RNNMode` [missing-attribute]
--> torch/backends/cudnn/rnn.py:21:20
|
21 | return int(_cudnn.RNNMode.gru)
| ^^^^^^^^^^^^^^
|
ERROR Unexpected keyword argument `self_ty` in function `torch._C._VariableFunctions._cudnn_init_dropout_state` [unexpected-keyword]
--> torch/backends/cudnn/rnn.py:59:21
|
59 | self_ty=torch.uint8,
| ^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `getVersionInt` [missing-attribute]
--> torch/backends/cusparselt/__init__.py:26:36
|
26 | __cusparselt_version = _cusparselt.getVersionInt()
| ^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Returned type `Literal['auto'] | None` is not assignable to declared return type `str` [bad-return]
--> torch/backends/opt_einsum/__init__.py:73:12
|
73 | return strategy
| ^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `add_argument` [missing-attribute]
--> torch/backends/xeon/run_cpu.py:838:5
|
838 | parser.add_argument(
| ^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `add_argument` [missing-attribute]
--> torch/backends/xeon/run_cpu.py:846:5
|
846 | parser.add_argument(
| ^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `add_argument` [missing-attribute]
--> torch/backends/xeon/run_cpu.py:856:5
|
856 | parser.add_argument(
| ^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `add_argument` [missing-attribute]
--> torch/backends/xeon/run_cpu.py:870:5
|
870 | parser.add_argument(
| ^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `add_argument` [missing-attribute]
--> torch/backends/xeon/run_cpu.py:878:5
|
878 | parser.add_argument("program_args", nargs=REMAINDER)
| ^^^^^^^^^^^^^^^^^^^
|
WARN `autocast` is deprecated [deprecated]
--> torch/cpu/amp/__init__.py:1:28
|
1 | from .autocast_mode import autocast
| --------
|
ERROR `list[bool] | None` is not subscriptable [unsupported-operation]
--> torch/profiler/_memory_profiler.py:233:17
|
233 | mutable[i] |= getattr(arg.alias_info, "is_write", False)
| ^^^^^^^^^^
|
ERROR Cannot set item in `None` [unsupported-operation]
--> torch/profiler/_memory_profiler.py:233:17
|
233 | mutable[i] |= getattr(arg.alias_info, "is_write", False)
| ^^^^^^^^^^
|
Object of class `NoneType` has no attribute `__setitem__`
ERROR `dict[tuple[int, device], Literal[True]] | dict[tuple[int, device], bool]` is not assignable to `dict[tuple[int, device], bool]` (caused by inconsistent types when breaking cycles) [bad-assignment]
--> torch/profiler/_memory_profiler.py:674:9
|
674 | / for event in self._op_tree.dfs():
675 | | if event.typed[0] == _EventType.Allocation:
676 | | alloc_fields = event.typed[1]
677 | | alloc_size = alloc_fields.alloc_size
678 | | is_allocation = alloc_size > 0
679 | | t = event.start_time_ns
| |________________________________________^
|
ERROR `in` is not supported between `int` and `Sized` [unsupported-operation]
--> torch/profiler/_memory_profiler.py:775:24
|
775 | or key.id in depends_on_gradient
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `Sized` has no attribute `update` [missing-attribute]
--> torch/profiler/_memory_profiler.py:779:21
|
779 | depends_on_gradient.update(ids)
| ^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Object of class `Sized` has no attribute `update` [missing-attribute]
--> torch/profiler/_memory_profiler.py:780:21
|
780 | depends_on_gradient.update(key.id for key in node.outputs)
| ^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Returned type `Sized | set[int]` is not assignable to declared return type `set[int]` [bad-return]
--> torch/profiler/_memory_profiler.py:788:24
|
788 | return depends_on_gradient
| ^^^^^^^^^^^^^^^^^^^
|
ERROR Argument `tuple[Unknown, auto, Unknown, Unknown]` is not assignable to parameter `object` with type `tuple[int, int, int, int]` in function `list.append` [bad-argument-type]
--> torch/profiler/_memory_profiler.py:1084:21
|
1084 | / (
1085 | | t,
1086 | | _ACTION_TO_INDEX[action],
1087 | | numbytes,
1088 | | get_category_index(key, version),
1089 | | )
| |_____________________^
|
ERROR Argument `tuple[Unknown, auto, Unknown, Unknown]` is not assignable to parameter `object` with type `tuple[int, int, int, int]` in function `list.append` [bad-argument-type]
--> torch/profiler/_memory_profiler.py:1094:21
|
1094 | / (
1095 | | t,
1096 | | _ACTION_TO_INDEX[action],
1097 | | -numbytes,
1098 | | get_category_index(key, version),
1099 | | )
| |_____________________^
|
ERROR Argument `tuple[Unknown, auto, Unknown, Unknown]` is not assignable to parameter `object` with type `tuple[int, int, int, int]` in function `list.append` [bad-argument-type]
--> torch/profiler/_memory_profiler.py:1102:21
|
1102 | / (
1103 | | t,
1104 | | _ACTION_TO_INDEX[action],
1105 | | numbytes,
1106 | | get_category_index(key, version + 1),
1107 | | )
| |_____________________^
|
ERROR Argument `tuple[Unknown, auto, Unknown, Unknown]` is not assignable to parameter `object` with type `tuple[int, int, int, int]` in function `list.append` [bad-argument-type]
--> torch/profiler/_memory_profiler.py:1112:21
|
1112 | / (
1113 | | t,
1114 | | _ACTION_TO_INDEX[action],
1115 | | -numbytes,
1116 | | get_category_index(key, version),
1117 | | )
| |_____________________^
|
ERROR Object of class `_KinetoEvent` has no attribute `duration_us` [missing-attribute]
--> torch/profiler/_utils.py:214:48
|
214 | end_time = (event.start_us() + event.duration_us()) * 1000
| ^^^^^^^^^^^^^^^^^
|
ERROR `Literal['cuda']` is not assignable to attribute `use_device` with type `None` [bad-assignment]
--> torch/profiler/profiler.py:164:31
|
164 | self.use_device = "cuda"
| ^^^^^^
|
ERROR `Literal['xpu']` is not assignable to attribute `use_device` with type `None` [bad-assignment]
--> torch/profiler/profiler.py:166:31
|
166 | self.use_device = "xpu"
| ^^^^^
|
ERROR `Literal['mtia']` is not assignable to attribute `use_device` with type `None` [bad-assignment]
--> torch/profiler/profiler.py:168:31
|
168 | self.use_device = "mtia"
| ^^^^^^
|
ERROR `Literal['hpu']` is not assignable to attribute `use_device` with type `None` [bad-assignment]
--> torch/profiler/profiler.py:170:31
|
170 | self.use_device = "hpu"
| ^^^^^
|
ERROR `str` is not assignable to attribute `use_device` with type `None` [bad-assignment]
--> torch/profiler/profiler.py:172:31
|
172 | self.use_device = _get_privateuse1_backend_name()
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Cannot set item in `dict[str, Backend | int | list[dict[str, Any]]]` [unsupported-operation]
--> torch/profiler/profiler.py:383:41
|
383 | dist_info["nccl_version"] = ".".join(str(v) for v in nccl_version)
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
Argument `str` is not assignable to parameter `value` with type `Backend | int | list[dict[str, Any]]` in function `dict.__setitem__`
ERROR No matching overload found for function `typing.MutableMapping.update` [no-matching-overload]
--> torch/sparse/__init__.py:626:33
|
626 | d.update(
| _________________________________^
627 | | indices=obj._indices(), is_coalesced=obj.is_coalesced()
628 | | )
| |_________________________^
|
Possible overloads:
(m: SupportsKeysAndGetItem[str, Size | layout], /) -> None
(m: SupportsKeysAndGetItem[str, Size | layout], /, **kwargs: Size | layout) -> None
(m: Iterable[tuple[str, Size | layout]], /) -> None
(m: Iterable[tuple[str, Size | layout]], /, **kwargs: Size | layout) -> None
(**kwargs: Size | layout) -> None [closest match]
ERROR No matching overload found for function `typing.MutableMapping.update` [no-matching-overload]
--> torch/sparse/__init__.py:631:33
|
631 | d.update(
| _________________________________^
632 | | compressed_indices=obj.crow_indices(),
633 | | plain_indices=obj.col_indices(),
634 | | )
| |_________________________^
|
Possible overloads:
(m: SupportsKeysAndGetItem[str, Size | layout], /) -> None
(m: SupportsKeysAndGetItem[str, Size | layout], /, **kwargs: Size | layout) -> None
(m: Iterable[tuple[str, Size | layout]], /) -> None
(m: Iterable[tuple[str, Size | layout]], /, **kwargs: Size | layout) -> None
(**kwargs: Size | layout) -> None [closest match]
ERROR No matching overload found for function `typing.MutableMapping.update` [no-matching-overload]
--> torch/sparse/__init__.py:637:33
|
637 | d.update(
| _________________________________^
638 | | compressed_indices=obj.ccol_indices(),
639 | | plain_indices=obj.row_indices(),
640 | | )
| |_________________________^
|
Possible overloads:
(m: SupportsKeysAndGetItem[str, Size | layout], /) -> None
(m: SupportsKeysAndGetItem[str, Size | layout], /, **kwargs: Size | layout) -> None
(m: Iterable[tuple[str, Size | layout]], /) -> None
(m: Iterable[tuple[str, Size | layout]], /, **kwargs: Size | layout) -> None
(**kwargs: Size | layout) -> None [closest match]
ERROR `Tensor | Unknown` is not subscriptable [unsupported-operation]
--> torch/sparse/_triton_ops.py:388:21
|
388 | accumulators[r] += blocks[p] @ others[q]
| ^^^^^^^^^^^^^^^
|
ERROR `None` is not subscriptable [unsupported-operation]
--> torch/sparse/_triton_ops.py:1299:9
|
1299 | _bsr_strided_addmm_kernel[grid](
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Cannot index into `bool` [index-error]
--> torch/sparse/_triton_ops.py:1428:26
|
1428 | mask=mask_k[None, :],
| ^^^^^^^^^^^^^^^
|
Object of class `bool` has no attribute `__getitem__`
ERROR Cannot index into `bool` [index-error]
--> torch/sparse/_triton_ops.py:1436:26
|
1436 | mask=mask_k[:, None],
| ^^^^^^^^^^^^^^^
|
Object of class `bool` has no attribute `__getitem__`
ERROR Expected a callable, got `None` [not-callable]
--> torch/sparse/_triton_ops.py:1973:16
|
1973 | sdpa = sampled_addmm(
| ^^^^^^^^^^^^^
|
ERROR Expected a callable, got `None` [not-callable]
--> torch/sparse/_triton_ops.py:1984:16
|
1984 | sdpa = bsr_softmax(sdpa)
| ^^^^^^^^^^^
|
ERROR Expected a callable, got `None` [not-callable]
--> torch/sparse/_triton_ops.py:1986:16
|
1986 | sdpa = bsr_dense_mm(sdpa, value)
| ^^^^^^^^^^^^
|
ERROR Argument `str` is not assignable to parameter `object` with type `LiteralString` in function `list.append` [bad-argument-type]
--> torch/sparse/_triton_ops_meta.py:235:26
|
235 | data_part.append(" " + repr(op_key).replace("'", '"') + ": {")
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR Argument `Generator[str, None, None]` is not assignable to parameter `iterable` with type `Iterable[LiteralString]` in function `list.extend` [bad-argument-type]
--> torch/sparse/_triton_ops_meta.py:237:26
|
237 | data_part.extend(f" {key}: {op_data[key]}," for key in sorted(op_data))
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
ERROR `+=` is not supported between `None` and `int` [unsupported-operation]
--> torch/sparse/_triton_ops_meta.py:370:21
|
370 | pbar.total += i + 1
| ^^^^^^^^^^^^^^^^^^^
|
Argument `None` is not assignable to parameter `value` with type `int` in function `int.__radd__`
WARN `assert_allclose` is deprecated [deprecated]
--> torch/testing/__init__.py:4:26
|
4 | from ._comparison import assert_allclose, assert_close as assert_close
| ---------------
|
ERROR Argument `bool | complex | float | int` is not assignable to parameter `x` with type `SupportsAbs[int]` in function `abs` [bad-argument-type]
--> torch/testing/_comparison.py:244:66
|
244 | rel_diff = float("inf") if expected == 0 else abs_diff / abs(expected)
| ^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `bool_` [missing-attribute]
--> torch/testing/_comparison.py:487:24
|
487 | cls.append(np.bool_)
| ^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `bool_` [missing-attribute]
--> torch/testing/_comparison.py:502:36
|
502 | elif isinstance(bool_like, np.bool_):
| ^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `number` [missing-attribute]
--> torch/testing/_comparison.py:581:24
|
581 | cls.append(np.number)
| ^^^^^^^^^
|
ERROR Object of class `NoneType` has no attribute `number` [missing-attribute]
--> torch/testing/_comparison.py:596:50
|
596 | if HAS_NUMPY and isinstance(number_like, np.number):
| ^^^^^^^^^
|
ERROR Function declared to return `list[Pair]`, but one or more paths are missing an explicit `return` [bad-return]
--> torch/testing/_comparison.py:1118:6
|
1118 | ) -> list[Pair]:
| ^^^^^^^^^^
|
ERROR `list[list[ErrorMeta]]` is not assignable to variable `error_metas` with type `list[ErrorMeta]` [bad-assignment]
--> torch/testing/_comparison.py:1313:19
|
1313 | error_metas = [error_metas]
| ^^^^^^^^^^^^^
|
ERROR Returned type `ErrorMeta` is not assignable to declared return type `list[ErrorMeta]` [bad-return]
--> torch/testing/_comparison.py:1314:12
|
1314 | return error_metas.pop()
| ^^^^^^^^^^^^^^^^^
|
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment