[CodeStyle] black -> ruff format migration - part 30

.pre-commit-config.yaml

@@ -85,7 +85,7 @@ repos:
 
             # | python/paddle/distributed/[g-z].+
 
-            # | python/paddle/[e-i].+
+            | python/paddle/[e-i].+
 
             # | python/paddle/j.+
 
@@ -141,7 +141,7 @@ repos:
 
             | python/paddle/distributed/[g-z].+
 
-            | python/paddle/[e-i].+
+            # | python/paddle/[e-i].+
 
             | python/paddle/j.+
 

python/paddle/framework/io.py

@@ -235,9 +235,9 @@ def _load_state_dict_from_save_inference_model(model_path, config):
                 structured_name = extra_var_info[var_name].get(
                     'structured_name', None
                 )
-                assert (
-                    structured_name is not None
-                ), f"Cannot find saved variable ({var_name})'s structured name in saved model."
+                assert structured_name is not None, (
+                    f"Cannot find saved variable ({var_name})'s structured name in saved model."
+                )
                 structured_para_dict[structured_name] = load_param_dict[
                     var_name
                 ]

python/paddle/hapi/model.py

@@ -368,13 +368,13 @@ def mode(self, value):
         self.model.mode = value
 
     def train_batch(self, inputs, labels=None, update=True):
-        assert (
-            self.model._optimizer
-        ), "model not ready, please call `model.prepare()` first"
+        assert self.model._optimizer, (
+            "model not ready, please call `model.prepare()` first"
+        )
         self.mode = 'train'
-        assert (
-            update is True
-        ), "Does not support `update == False` in static graph mode by now."
+        assert update is True, (
+            "Does not support `update == False` in static graph mode by now."
+        )
         return self._run(inputs, labels)
 
     def eval_batch(self, inputs, labels=None):
@@ -500,16 +500,16 @@ def _load_optimizer(self, state, executor):
                 # However, dygraph wouldn't save it.
                 if var.name not in state:
                     continue
-            assert (
-                var.name in converted_state
-            ), f"variable [{var.name}] is not in optimizer state file"
+            assert var.name in converted_state, (
+                f"variable [{var.name}] is not in optimizer state file"
+            )
             self._set_var(var.name, converted_state[var.name])
 
     def _run(self, inputs, labels=None):
         compiled_prog = self._compiled_progs.get(self.mode, None)
-        assert (
-            compiled_prog
-        ), "Model is not ready, please call `model.prepare()` first"
+        assert compiled_prog, (
+            "Model is not ready, please call `model.prepare()` first"
+        )
 
         inputs = to_list(inputs)
         if labels is not None:
@@ -689,9 +689,9 @@ def _make_program(self, mode):
         }
 
     def _initialize(self, prog, mode):
-        assert (
-            self.model._place is not None
-        ), "device is not set, please call `model.prepare()` first"
+        assert self.model._place is not None, (
+            "device is not set, please call `model.prepare()` first"
+        )
 
         place = self.model._place
 
@@ -756,13 +756,13 @@ def mode(self, value):
         self.model.mode = value
 
     def train_batch(self, inputs, labels=None, update=True):
-        assert (
-            self.model._optimizer
-        ), "model not ready, please call `model.prepare()` first"
+        assert self.model._optimizer, (
+            "model not ready, please call `model.prepare()` first"
+        )
         self.mode = 'train'
-        assert (
-            update is True
-        ), "Does not support `update == False` in static graph mode by now."
+        assert update is True, (
+            "Does not support `update == False` in static graph mode by now."
+        )
         return self._run(inputs, labels)
 
     def eval_batch(self, inputs, labels=None):
@@ -919,9 +919,9 @@ def _load_optimizer(self, state, executor):
                                 converted_state.pop(dy_state_name)
                             )
 
-            assert (
-                var.name in converted_state
-            ), f"variable [{var.name}] is not in optimizer state file"
+            assert var.name in converted_state, (
+                f"variable [{var.name}] is not in optimizer state file"
+            )
             self._set_var(var, converted_state[var.name])
 
     def _set_var(self, var, ndarray):
@@ -940,9 +940,9 @@ def _set_var(self, var, ndarray):
 
     def _run(self, inputs, labels=None):
         compiled_prog = self._compiled_progs.get(self.mode, None)
-        assert (
-            compiled_prog
-        ), "Model is not ready, please call `model.prepare()` first"
+        assert compiled_prog, (
+            "Model is not ready, please call `model.prepare()` first"
+        )
 
         inputs = to_list(inputs)
         if labels is not None:
@@ -1141,9 +1141,9 @@ def _compile_and_initialize(self, prog, mode):
         if compiled_prog is not None:
             return compiled_prog
 
-        assert (
-            self.model._place is not None
-        ), "device is not set, please call `model.prepare()` first"
+        assert self.model._place is not None, (
+            "device is not set, please call `model.prepare()` first"
+        )
 
         place = self.model._place
 
@@ -1234,9 +1234,9 @@ def mode(self, value):
 
     # TODO multi device in dygraph mode not implemented at present time
     def train_batch(self, inputs, labels=None, update=True):
-        assert (
-            self.model._optimizer
-        ), "model not ready, please call `model.prepare()` first"
+        assert self.model._optimizer, (
+            "model not ready, please call `model.prepare()` first"
+        )
         self.model.network.train()
         self.mode = 'train'
         inputs = to_list(inputs)
@@ -2031,7 +2031,9 @@ def _check_pure_fp16_configs():
                 assert isinstance(
                     self._optimizer._grad_clip,
                     (paddle.nn.ClipGradByGlobalNorm, paddle.nn.ClipGradByNorm),
-                ), "Only ClipGradByNorm and ClipGradByGlobalNorm are supported in amp training with level=O2 currently."
+                ), (
+                    "Only ClipGradByNorm and ClipGradByGlobalNorm are supported in amp training with level=O2 currently."
+                )
 
         self._adapter._amp_custom_lists = {}
         self._adapter._amp_configs = {}
@@ -2188,9 +2190,9 @@ def prepare(
 
         metrics = metrics or []
         for metric in to_list(metrics):
-            assert isinstance(
-                metric, Metric
-            ), f"{metric.__class__.__name__} is not sub class of Metric"
+            assert isinstance(metric, Metric), (
+                f"{metric.__class__.__name__} is not sub class of Metric"
+            )
         self._metrics = to_list(metrics)
         self._prepare_amp(amp_configs)
 
@@ -2353,9 +2355,9 @@ def fit(
         if isinstance(batch_size, (tuple, list)) and all(
             isinstance(x, int) for x in batch_size
         ):
-            assert (
-                len(batch_size) == 2
-            ), "batch_size length error, expected train_batch_size and eval_batch_size."
+            assert len(batch_size) == 2, (
+                "batch_size length error, expected train_batch_size and eval_batch_size."
+            )
             train_batch_size, eval_batch_size = batch_size
         elif isinstance(batch_size, int):
             train_batch_size, eval_batch_size = batch_size, batch_size
@@ -2748,9 +2750,9 @@ def _save_inference_model(self, path: str) -> None:
             params_filename = file_prefix + INFER_PARAMS_SUFFIX
 
             prog = self._adapter._progs.get('test', None)
-            assert (
-                prog
-            ), "Model is not ready, please call `model.prepare()` first"
+            assert prog, (
+                "Model is not ready, please call `model.prepare()` first"
+            )
 
             if in_pir_mode():
                 infer_prog = prog
@@ -2914,9 +2916,9 @@ def summary(
                 {'total_params': 61610, 'trainable_params': 61610}
 
         """
-        assert (
-            input_size is not None or self._inputs is not None
-        ), "'input_size' or 'self._input' must be set"
+        assert input_size is not None or self._inputs is not None, (
+            "'input_size' or 'self._input' must be set"
+        )
         if input_size is not None:
             _input_size = input_size
         else:

python/paddle/hapi/model_summary.py

@@ -348,10 +348,10 @@ def summary(
         for item in input_size:
             if isinstance(item, int):
                 item = (item,)
-            assert isinstance(
-                item, (tuple, InputSpec)
-            ), f'When input_size is list, \
+            assert isinstance(item, (tuple, InputSpec)), (
+                f'When input_size is list, \
             expect item in input_size is a tuple or InputSpec, but got {type(item)}'
+            )
 
             if isinstance(item, InputSpec):
                 _input_size.append(tuple(item.shape))

python/paddle/incubate/asp/asp.py

@@ -464,9 +464,9 @@ def prune_model(
         'mask_2d_greedy': MaskAlgo.MASK_2D_GREEDY,
         'mask_2d_best': MaskAlgo.MASK_2D_BEST,
     }
-    assert (
-        mask_algo in MaskAlgo_mapping
-    ), 'The "mask_algo" should be one of ["mask_1d", "mask_2d_greedy", "mask_2d_best"]'
+    assert mask_algo in MaskAlgo_mapping, (
+        'The "mask_algo" should be one of ["mask_1d", "mask_2d_greedy", "mask_2d_best"]'
+    )
 
     prune_func = None
     if isinstance(model, paddle.nn.Layer):
@@ -685,9 +685,9 @@ def prune_model_by_layer(
             target_program = None
             for param in layer.parameters():
                 target_program = param.block.program
-            assert (
-                target_program is not None
-            ), 'Cannot get paddle.static.Program from Paddle.nn.Layer.'
+            assert target_program is not None, (
+                'Cannot get paddle.static.Program from Paddle.nn.Layer.'
+            )
             return ASPHelper.prune_model_by_program(
                 place,
                 target_program,
@@ -795,7 +795,9 @@ def _is_supported_layer(
         return False
 
     @classmethod
-    def _get_prune_func_by_name(cls, param_name: str) -> Callable[
+    def _get_prune_func_by_name(
+        cls, param_name: str
+    ) -> Callable[
         [npt.NDArray[Any], int, int, MaskAlgo, str],
         tuple[npt.NDArray[Any], npt.NDArray[Any]],
     ]:
@@ -1036,9 +1038,9 @@ def set_state_dict(self, state_dict: dict[str, Tensor]) -> None:
         )
         for param_name, var in asp_info.mask_vars.items():
             param_mask_name = ASPHelper._get_mask_name(param_name)
-            assert (
-                param_mask_name in state_dict
-            ), f"The {param_mask_name} is not found."
+            assert param_mask_name in state_dict, (
+                f"The {param_mask_name} is not found."
+            )
             var.set_value(state_dict[param_mask_name])
             asp_info.update_masks(param_name, var.numpy())
         return self._optimizer.set_state_dict(state_dict)

python/paddle/incubate/asp/utils.py

@@ -74,9 +74,9 @@ def get_checking_method(mask_algo: MaskAlgo) -> CheckMethod:
                 >>> print(CheckMethod.get_checking_method(MaskAlgo.MASK_2D_BEST))
                 CheckMethod.CHECK_2D
         """
-        assert isinstance(
-            mask_algo, MaskAlgo
-        ), "mask_algo should be MaskAlgo type"
+        assert isinstance(mask_algo, MaskAlgo), (
+            "mask_algo should be MaskAlgo type"
+        )
         if mask_algo == MaskAlgo.MASK_1D:
             return CheckMethod.CHECK_1D
         else: