pytorch-ignite 0.6.0.dev20251215__tar.gz → 0.6.0.dev20251217__tar.gz

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/.github/workflows/typing-checks.yml

@@ -11,7 +11,7 @@ on:
       - "tests/ignite/**"
       - "requirements-dev.txt"
       - "pyproject.toml"
-      - "mypy.ini"
+      - "pyrefly.toml"
       - ".github/workflows/typing-checks.yml"
   pull_request:
     paths:
@@ -20,7 +20,7 @@ on:
       - "tests/ignite/**"
       - "requirements-dev.txt"
       - "pyproject.toml"
-      - "mypy.ini"
+      - "pyrefly.toml"
       - ".github/workflows/typing-checks.yml"
   workflow_dispatch:
 
@@ -30,7 +30,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  mypy:
+  pyrefly:
     runs-on: ubuntu-latest
     strategy:
       matrix:
@@ -60,14 +60,7 @@ jobs:
         run: uv pip install torch torchvision --index-url https://download.pytorch.org/whl/cpu
 
       - name: Install dependencies
-        run: |
-          uv pip install -r requirements-dev.txt
-          uv pip install .
-          uv pip install mypy
-          uv pip install pyrefly
-
-      - name: Run MyPy type checking
-        run: mypy
+        run: uv pip install -r requirements-dev.txt
 
       - name: Run Pyrefly type checking
         run: pyrefly check

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/CONTRIBUTING.md

@@ -40,7 +40,7 @@ into the following categories:
       - [Formatting the code with a pre-commit hook](#formatting-the-code-with-a-pre-commit-hook)
     - [Run tests](#run-tests)
       - [Run distributed tests only on CPU](#run-distributed-tests-only-on-cpu)
-    - [Run Mypy checks](#run-mypy-checks)
+    - [Run Pyrefly checks](#run-pyrefly-checks)
     - [Send a PR](#send-a-pr)
       - [Sync up with the upstream](#sync-up-with-the-upstream)
   - [Writing documentation](#writing-documentation)
@@ -214,15 +214,15 @@ export WORLD_SIZE=2
 CUDA_VISIBLE_DEVICES="" pytest --dist=each --tx $WORLD_SIZE*popen//python=python tests/ -m distributed -vvv
 ```
 
-#### Run Mypy checks:
+#### Run Pyrefly checks:
 
-To run mypy to check the optional static type:
+To run pyrefly to check the optional static type:
 
 ```bash
-mypy
+pyrefly check
 ```
 
-To change any config for specif folder, please see the file mypy.ini
+To change any config for specific folder, please see the file pyrefly.toml
 
 #### Send a PR
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pytorch-ignite
-Version: 0.6.0.dev20251215
+Version: 0.6.0.dev20251217
 Summary: A lightweight library to help with training neural networks in PyTorch.
 Project-URL: Homepage, https://pytorch-ignite.ai
 Project-URL: Repository, https://github.com/pytorch/ignite

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/conda.recipe/meta.yaml

@@ -1,4 +1,4 @@
-{% set __version__ = "0.6.0.dev20251215" %}
+{% set __version__ = "0.6.0.dev20251217" %}
 
 package:
   name: ignite

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/__init__.py

@@ -6,4 +6,4 @@ import ignite.handlers
 import ignite.metrics
 import ignite.utils
 
-__version__ = "0.6.0.dev20251215"
+__version__ = "0.6.0.dev20251217"

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/contrib/engines/tbptt.py

@@ -117,6 +117,5 @@ def create_supervised_tbptt_trainer(
         return sum(loss_list) / len(loss_list)
 
     engine = Engine(_update)
-    # pyrefly: ignore [bad-argument-type]
     engine.register_events(*Tbptt_Events)
     return engine

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/distributed/comp_models/horovod.py

@@ -1,6 +1,6 @@
 import os
 import warnings
-from typing import Any, Callable, cast, List, Mapping, Optional, Tuple
+from typing import Any, Callable, cast, List, Mapping, Optional, Tuple, TYPE_CHECKING
 
 import torch
 
@@ -20,6 +20,11 @@ try:
 except ImportError:
     has_hvd_support = False
 
+    if TYPE_CHECKING:
+        # Tell the type checker that hvd imports are always defined.
+        import horovod.torch as hvd
+        from horovod import run as hvd_mp_spawn
+
 
 if has_hvd_support:
     HOROVOD = "horovod"
@@ -171,11 +176,8 @@ if has_hvd_support:
             return group
 
         _reduce_op_map = {
-            # pyrefly: ignore [unbound-name]
             "SUM": hvd.mpi_ops.Sum,
-            # pyrefly: ignore [unbound-name]
             "AVERAGE": hvd.mpi_ops.Average,
-            # pyrefly: ignore [unbound-name]
             "ADASUM": hvd.mpi_ops.Adasum,
         }
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/distributed/launcher.py

@@ -322,19 +322,15 @@ class Parallel:
             idist.initialize(self.backend, init_method=self.init_method)
 
         # The logger can be setup from now since idist.initialize() has been called (if needed)
-        self._logger = setup_logger(__name__ + "." + self.__class__.__name__)  # type: ignore[assignment]
+        self._logger = setup_logger(__name__ + "." + self.__class__.__name__)
 
         if self.backend is not None:
             if self._spawn_params is None:
-                self._logger.info(  # type: ignore[attr-defined]
-                    f"Initialized processing group with backend: '{self.backend}'"
-                )
+                self._logger.info(f"Initialized processing group with backend: '{self.backend}'")
             else:
-                self._logger.info(  # type: ignore[attr-defined]
-                    f"Initialized distributed launcher with backend: '{self.backend}'"
-                )
+                self._logger.info(f"Initialized distributed launcher with backend: '{self.backend}'")
                 msg = "\n\t".join([f"{k}: {v}" for k, v in self._spawn_params.items() if v is not None])
-                self._logger.info(f"- Parameters to spawn processes: \n\t{msg}")  # type: ignore[attr-defined]
+                self._logger.info(f"- Parameters to spawn processes: \n\t{msg}")
 
         return self
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/engine/engine.py

@@ -148,12 +148,11 @@ class Engine(Serializable):
         self.should_interrupt = False
         self.state = State()
         self._state_dict_user_keys: List[str] = []
-        self._allowed_events: List[EventEnum] = []
+        self._allowed_events: List[Union[str, EventEnum]] = []
 
         self._dataloader_iter: Optional[Iterator[Any]] = None
         self._init_iter: Optional[int] = None
 
-        # pyrefly: ignore [bad-argument-type]
         self.register_events(*Events)
 
         if self._process_function is None:
@@ -164,9 +163,7 @@ class Engine(Serializable):
         # generator provided by self._internal_run_as_gen
         self._internal_run_generator: Optional[Generator[Any, None, State]] = None
 
-    def register_events(
-        self, *event_names: Union[List[str], List[EventEnum]], event_to_attr: Optional[dict] = None
-    ) -> None:
+    def register_events(self, *event_names: Union[str, EventEnum], event_to_attr: Optional[dict] = None) -> None:
         """Add events that can be fired.
 
         Registering an event will let the user trigger these events at any point.
@@ -451,7 +448,7 @@ class Engine(Serializable):
                 first, others = ((resolved_engine,), args[1:])
             else:
                 # metrics do not provide engine when registered
-                first, others = (tuple(), args)  # type: ignore[assignment]
+                first, others = (tuple(), args)
 
             func(*first, *(event_args + others), **kwargs)
 
@@ -990,9 +987,9 @@ class Engine(Serializable):
     def _internal_run_as_gen(self) -> Generator[Any, None, State]:
         self.should_terminate = self.should_terminate_single_epoch = self.should_interrupt = False
         self._init_timers(self.state)
+        start_time = time.time()
         try:
             try:
-                start_time = time.time()
                 self._fire_event(Events.STARTED)
                 yield from self._maybe_terminate_or_interrupt()
 
@@ -1011,7 +1008,7 @@ class Engine(Serializable):
                     # time is available for handlers but must be updated after fire
                     self.state.times[Events.EPOCH_COMPLETED.name] = epoch_time_taken
 
-                    if self.should_terminate_single_epoch != "skip_epoch_completed":  # type: ignore[comparison-overlap]
+                    if self.should_terminate_single_epoch != "skip_epoch_completed":
                         handlers_start_time = time.time()
                         self._fire_event(Events.EPOCH_COMPLETED)
                         epoch_time_taken += time.time() - handlers_start_time
@@ -1039,13 +1036,12 @@ class Engine(Serializable):
                     "https://github.com/pytorch/ignite/issues/new/choose"
                 )
 
-            # pyrefly: ignore [unbound-name]
             time_taken = time.time() - start_time
             # time is available for handlers but must be updated after fire
             self.state.times[Events.COMPLETED.name] = time_taken
 
             # do not fire Events.COMPLETED if we terminated the run with flag `skip_completed=True`
-            if self.should_terminate != "skip_completed":  # type: ignore[comparison-overlap]
+            if self.should_terminate != "skip_completed":
                 handlers_start_time = time.time()
                 self._fire_event(Events.COMPLETED)
                 time_taken += time.time() - handlers_start_time
@@ -1191,9 +1187,9 @@ class Engine(Serializable):
         # internal_run without generator for BC
         self.should_terminate = self.should_terminate_single_epoch = self.should_interrupt = False
         self._init_timers(self.state)
+        start_time = time.time()
         try:
             try:
-                start_time = time.time()
                 self._fire_event(Events.STARTED)
                 self._maybe_terminate_legacy()
 
@@ -1212,7 +1208,7 @@ class Engine(Serializable):
                     # time is available for handlers but must be updated after fire
                     self.state.times[Events.EPOCH_COMPLETED.name] = epoch_time_taken
 
-                    if self.should_terminate_single_epoch != "skip_epoch_completed":  # type: ignore[comparison-overlap]
+                    if self.should_terminate_single_epoch != "skip_epoch_completed":
                         handlers_start_time = time.time()
                         self._fire_event(Events.EPOCH_COMPLETED)
                         epoch_time_taken += time.time() - handlers_start_time
@@ -1240,13 +1236,12 @@ class Engine(Serializable):
                     "https://github.com/pytorch/ignite/issues/new/choose"
                 )
 
-            # pyrefly: ignore [unbound-name]
             time_taken = time.time() - start_time
             # time is available for handlers but must be updated after fire
             self.state.times[Events.COMPLETED.name] = time_taken
 
             # do not fire Events.COMPLETED if we terminated the run with flag `skip_completed=True`
-            if self.should_terminate != "skip_completed":  # type: ignore[comparison-overlap]
+            if self.should_terminate != "skip_completed":
                 handlers_start_time = time.time()
                 self._fire_event(Events.COMPLETED)
                 time_taken += time.time() - handlers_start_time

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/checkpoint.py

@@ -315,7 +315,7 @@ class Checkpoint(Serializable):
     """
 
     SAVED_CHECKPOINT = CheckpointEvents.SAVED_CHECKPOINT
-    Item = NamedTuple("Item", [("priority", int), ("filename", str)])
+    Item = NamedTuple("Item", [("priority", Union[int, float]), ("filename", str)])
     _state_dict_all_req_keys = ("_saved",)
 
     def __init__(
@@ -323,7 +323,7 @@ class Checkpoint(Serializable):
         to_save: Mapping,
         save_handler: Union[str, Path, Callable, BaseSaveHandler],
         filename_prefix: str = "",
-        score_function: Optional[Callable] = None,
+        score_function: Optional[Callable[[Engine], Union[int, float]]] = None,
         score_name: Optional[str] = None,
         n_saved: Union[int, None] = 1,
         global_step_transform: Optional[Callable] = None,
@@ -440,7 +440,6 @@ class Checkpoint(Serializable):
 
     def __call__(self, engine: Engine) -> None:
         if not engine.has_registered_events(CheckpointEvents.SAVED_CHECKPOINT):
-            # pyrefly: ignore [bad-argument-type]
             engine.register_events(*CheckpointEvents)
         global_step = None
         if self.global_step_transform is not None:
@@ -455,7 +454,6 @@ class Checkpoint(Serializable):
                 global_step = engine.state.get_event_attrib_value(Events.ITERATION_COMPLETED)
             priority = global_step
 
-        # pyrefly: ignore [bad-argument-type]
         if self._check_lt_n_saved() or self._compare_fn(priority):
             priority_str = f"{priority}" if isinstance(priority, numbers.Integral) else f"{priority:.4f}"
 
@@ -497,7 +495,6 @@ class Checkpoint(Serializable):
                 if isinstance(self.save_handler, BaseSaveHandler):
                     self.save_handler.remove(item.filename)
 
-            # pyrefly: ignore [bad-argument-type]
             self._saved.append(Checkpoint.Item(priority, filename))
             self._saved.sort(key=lambda it: it[0])
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/clearml_logger.py

@@ -862,7 +862,7 @@ class ClearMLSaver(DiskSaver):
         except ImportError:
             try:
                 # Backwards-compatibility for legacy Trains SDK
-                from trains import Task  # type: ignore[no-redef]
+                from trains import Task
             except ImportError:
                 raise ModuleNotFoundError(
                     "This contrib module requires clearml to be installed. "
@@ -937,7 +937,7 @@ class ClearMLSaver(DiskSaver):
         except ImportError:
             try:
                 # Backwards-compatibility for legacy Trains SDK
-                from trains.binding.frameworks import WeightsFileHandler  # type: ignore[no-redef]
+                from trains.binding.frameworks import WeightsFileHandler
             except ImportError:
                 raise ModuleNotFoundError(
                     "This contrib module requires clearml to be installed. "

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/lr_finder.py

@@ -98,16 +98,18 @@ class FastaiLRFinder:
         self._best_loss = None
         self._diverge_flag = False
 
+        assert trainer.state.epoch_length is not None
+        assert trainer.state.max_epochs is not None
+
         # attach LRScheduler to trainer.
         if num_iter is None:
-            # pyrefly: ignore [unsupported-operation]
             num_iter = trainer.state.epoch_length * trainer.state.max_epochs
         else:
-            max_iter = trainer.state.epoch_length * trainer.state.max_epochs  # type: ignore[operator]
+            max_iter = trainer.state.epoch_length * trainer.state.max_epochs
             if max_iter < num_iter:
                 max_iter = num_iter
                 trainer.state.max_iters = num_iter
-                trainer.state.max_epochs = ceil(num_iter / trainer.state.epoch_length)  # type: ignore[operator]
+                trainer.state.max_epochs = ceil(num_iter / trainer.state.epoch_length)
 
         if not trainer.has_event_handler(self._reached_num_iterations):
             trainer.add_event_handler(Events.ITERATION_COMPLETED, self._reached_num_iterations, num_iter)
@@ -179,18 +181,14 @@ class FastaiLRFinder:
         loss = idist.all_reduce(loss)
         lr = self._lr_schedule.get_param()
         self._history["lr"].append(lr)
-        if trainer.state.iteration == 1:
-            self._best_loss = loss  # type: ignore[assignment]
-        else:
-            if smooth_f > 0:
-                loss = smooth_f * loss + (1 - smooth_f) * self._history["loss"][-1]
-            # pyrefly: ignore [unsupported-operation]
-            if loss < self._best_loss:
-                self._best_loss = loss
+        if trainer.state.iteration != 1 and smooth_f > 0:
+            loss = smooth_f * loss + (1 - smooth_f) * self._history["loss"][-1]
+        if self._best_loss is None or loss < self._best_loss:
+            self._best_loss = loss
         self._history["loss"].append(loss)
 
         # Check if the loss has diverged; if it has, stop the trainer
-        if self._history["loss"][-1] > diverge_th * self._best_loss:  # type: ignore[operator]
+        if self._history["loss"][-1] > diverge_th * self._best_loss:
             self._diverge_flag = True
             self.logger.info("Stopping early, the loss has diverged")
             trainer.terminate()

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/param_scheduler.py

@@ -1122,13 +1122,14 @@ def create_lr_scheduler_with_warmup(
             f"but given {type(lr_scheduler)}"
         )
 
-    if not isinstance(warmup_duration, numbers.Integral):
+    if not isinstance(warmup_duration, int):
         raise TypeError(f"Argument warmup_duration should be integer, but given {warmup_duration}")
 
     if not (warmup_duration > 1):
         raise ValueError(f"Argument warmup_duration should be at least 2 events, but given {warmup_duration}")
 
     warmup_schedulers: List[ParamScheduler] = []
+    milestones_values: List[Tuple[int, float]] = []
 
     for param_group_index, param_group in enumerate(lr_scheduler.optimizer.param_groups):
         if warmup_end_value is None:
@@ -1154,7 +1155,6 @@ def create_lr_scheduler_with_warmup(
             init_lr = lr_scheduler.get_param()
             if init_lr == param_group_warmup_end_value:
                 if warmup_duration > 2:
-                    # pyrefly: ignore [unsupported-operation]
                     d = (param_group_warmup_end_value - warmup_start_value) / (warmup_duration - 1)
                     milestones_values[-1] = (warmup_duration - 2, param_group_warmup_end_value - d)
                 else:
@@ -1164,7 +1164,6 @@ def create_lr_scheduler_with_warmup(
             PiecewiseLinear(
                 lr_scheduler.optimizer,
                 param_name="lr",
-                # pyrefly: ignore [bad-argument-type]
                 milestones_values=milestones_values,
                 param_group_index=param_group_index,
                 save_history=save_history,
@@ -1177,7 +1176,6 @@ def create_lr_scheduler_with_warmup(
         warmup_scheduler,
         lr_scheduler,
     ]
-    # pyrefly: ignore [unbound-name, unsupported-operation]
     durations = [milestones_values[-1][0] + 1]
     # pyrefly: ignore [bad-argument-type]
     combined_scheduler = ConcatScheduler(schedulers, durations=durations, save_history=save_history)
@@ -1655,13 +1653,13 @@ class ReduceLROnPlateauScheduler(ParamScheduler):
         self.trainer = trainer
         self.optimizer = optimizer
 
+        min_lr: Union[float, List[float]]
         if "min_lr" in scheduler_kwargs and param_group_index is not None:
             min_lr = scheduler_kwargs["min_lr"]
             if not isinstance(min_lr, float):
                 raise TypeError(f"When param_group_index is given, min_lr should be a float, but given {type(min_lr)}")
             _min_lr = min_lr
             min_lr = [0] * len(optimizer.param_groups)
-            # pyrefly: ignore [unsupported-operation]
             min_lr[param_group_index] = _min_lr
         else:
             min_lr = 0
@@ -1676,11 +1674,11 @@ class ReduceLROnPlateauScheduler(ParamScheduler):
             _scheduler_kwargs["verbose"] = False
 
         self.scheduler = ReduceLROnPlateau(optimizer, **_scheduler_kwargs)
-        self.scheduler._reduce_lr = self._reduce_lr  # type: ignore[method-assign]
+        self.scheduler._reduce_lr = self._reduce_lr
 
         self._state_attrs += ["metric_name", "scheduler"]
 
-    def __call__(self, engine: Engine, name: Optional[str] = None) -> None:  # type: ignore[override]
+    def __call__(self, engine: Engine, name: Optional[str] = None) -> None:
         if not hasattr(engine.state, "metrics") or self.metric_name not in engine.state.metrics:
             raise ValueError(
                 "Argument engine should have in its 'state', attribute 'metrics' "

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/state_param_scheduler.py

@@ -1,7 +1,7 @@
 import numbers
 import warnings
 from bisect import bisect_right
-from typing import Any, List, Sequence, Tuple, Union
+from typing import Any, Callable, List, Sequence, Tuple, Union
 
 from ignite.engine import CallableEventWithFilter, Engine, Events, EventsList
 from ignite.handlers.param_scheduler import BaseParamScheduler
@@ -183,7 +183,13 @@ class LambdaStateScheduler(StateParamScheduler):
 
     """
 
-    def __init__(self, lambda_obj: Any, param_name: str, save_history: bool = False, create_new: bool = False):
+    def __init__(
+        self,
+        lambda_obj: Callable[[int], Union[List[float], float]],
+        param_name: str,
+        save_history: bool = False,
+        create_new: bool = False,
+    ):
         super(LambdaStateScheduler, self).__init__(param_name, save_history, create_new)
 
         if not callable(lambda_obj):
@@ -193,7 +199,6 @@ class LambdaStateScheduler(StateParamScheduler):
         self._state_attrs += ["lambda_obj"]
 
     def get_param(self) -> Union[List[float], float]:
-        # pyrefly: ignore [bad-return]
         return self.lambda_obj(self.event_index)
 
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/time_profilers.py

@@ -500,14 +500,14 @@ class HandlersTimeProfiler:
 
         self.dataflow_times: List[float] = []
         self.processing_times: List[float] = []
-        self.event_handlers_times: Dict[EventEnum, Dict[str, List[float]]] = {}
+        self.event_handlers_times: Dict[Union[str, EventEnum], Dict[str, List[float]]] = {}
 
     @staticmethod
     def _get_callable_name(handler: Callable) -> str:
         # get name of the callable handler
         return getattr(handler, "__qualname__", handler.__class__.__name__)
 
-    def _create_wrapped_handler(self, handler: Callable, event: EventEnum) -> Callable:
+    def _create_wrapped_handler(self, handler: Callable, event: Union[str, EventEnum]) -> Callable:
         @functools.wraps(handler)
         def _timeit_handler(*args: Any, **kwargs: Any) -> None:
             self._event_handlers_timer.reset()
@@ -532,7 +532,7 @@ class HandlersTimeProfiler:
         t = self._dataflow_timer.value()
         self.dataflow_times.append(t)
 
-    def _reset(self, event_handlers_names: Mapping[EventEnum, List[str]]) -> None:
+    def _reset(self, event_handlers_names: Mapping[Union[str, EventEnum], List[str]]) -> None:
         # reset the variables used for profiling
         self.dataflow_times = []
         self.processing_times = []

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/tqdm_logger.py

@@ -223,7 +223,7 @@ class ProgressBar(BaseLogger):
         super(ProgressBar, self).attach(engine, log_handler, event_name)
         engine.add_event_handler(closing_event_name, self._close)
 
-    def attach_opt_params_handler(  # type: ignore[empty-body]
+    def attach_opt_params_handler(
         self,
         engine: Engine,
         event_name: Union[str, Events],

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/handlers/visdom_logger.py

@@ -1,7 +1,7 @@
 """Visdom logger and its helper handlers."""
 
 import os
-from typing import Any, Callable, Dict, List, Optional, Union
+from typing import Any, Callable, Dict, List, Optional, Union, TYPE_CHECKING
 
 import torch
 import torch.nn as nn
@@ -165,7 +165,7 @@ class VisdomLogger(BaseLogger):
                 "pip install git+https://github.com/fossasia/visdom.git"
             )
 
-        if num_workers > 0:
+        if num_workers > 0 or TYPE_CHECKING:
             # If visdom is installed, one of its dependencies `tornado`
             # requires also `futures` to be installed.
             # Let's check anyway if we can import it.
@@ -199,7 +199,6 @@ class VisdomLogger(BaseLogger):
 
         self.executor: Union[_DummyExecutor, "ThreadPoolExecutor"] = _DummyExecutor()
         if num_workers > 0:
-            # pyrefly: ignore [unbound-name]
             self.executor = ThreadPoolExecutor(max_workers=num_workers)
 
     def _save(self) -> None:

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/metrics/accuracy.py

@@ -254,10 +254,10 @@ class Accuracy(_BaseClassification):
             y_pred = torch.transpose(y_pred, 1, last_dim - 1).reshape(-1, num_classes)
             y = torch.transpose(y, 1, last_dim - 1).reshape(-1, num_classes)
             correct = torch.all(y == y_pred.type_as(y), dim=-1)
+        else:
+            raise ValueError(f"Unexpected type: {self._type}")
 
-        # pyrefly: ignore [unbound-name]
         self._num_correct += torch.sum(correct).to(self._device)
-        # pyrefly: ignore [unbound-name]
         self._num_examples += correct.shape[0]
 
     @sync_all_reduce("_num_examples", "_num_correct")

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/ignite/metrics/nlp/rouge.py

@@ -1,6 +1,5 @@
 from abc import ABCMeta, abstractmethod
-from collections import namedtuple
-from typing import Any, Callable, List, Mapping, Optional, Sequence, Tuple, Union
+from typing import Any, Callable, List, Mapping, NamedTuple, Optional, Sequence, Tuple, Union
 
 import torch
 
@@ -13,24 +12,25 @@ from ignite.metrics.nlp.utils import lcs, ngrams
 __all__ = ["Rouge", "RougeN", "RougeL"]
 
 
-# pyrefly: ignore [invalid-inheritance]
-class Score(namedtuple("Score", ["match", "candidate", "reference"])):
+class Score(NamedTuple):
     r"""
     Computes precision and recall for given matches, candidate and reference lengths.
     """
 
+    match: int
+    candidate: int
+    reference: int
+
     def precision(self) -> float:
         """
         Calculates precision.
         """
-        # pyrefly: ignore [missing-attribute]
         return self.match / self.candidate if self.candidate > 0 else 0
 
     def recall(self) -> float:
         """
         Calculates recall.
         """
-        # pyrefly: ignore [missing-attribute]
         return self.match / self.reference if self.reference > 0 else 0
 
 

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/requirements-dev.txt

@@ -1,7 +1,7 @@
 # Tests
 dill
 filelock
-mypy
+pyrefly
 numpy
 pre-commit
 pytest

{pytorch_ignite-0.6.0.dev20251215 → pytorch_ignite-0.6.0.dev20251217}/tests/ignite/handlers/test_ema_handler.py

@@ -97,7 +97,7 @@ def test_ema_warmup_func(get_dummy_model):
 def test_ema_invalid_model():
     with pytest.raises(ValueError, match="model should be an instance of nn.Module or its subclasses"):
         model = "Invalid Model"
-        EMAHandler(model)  # type: ignore
+        EMAHandler(model)
 
 
 @pytest.mark.distributed