comments

examples/llm_compression/openvino/tiny_llama_find_hyperparams/main.py

@@ -132,7 +132,7 @@ def print_results(optimized_model: ov.Model, similarity: float) -> None:
     else:
         print(best_params_info)
     footprint = Path(MODEL_PATH).with_suffix(".bin").stat().st_size
-    print(f"Memory footprint: {footprint / 2**20:.2f} MB")
+    print(f"Memory footprint: {footprint / 2**20 :.2f} MB")
     print(f"Similarity: {similarity:.2f}")
 
 

examples/torch/common/utils.py

@@ -166,7 +166,9 @@ def is_staged_quantization(config):
     if isinstance(compression_config, list):
         compression_config = compression_config[0]
     algo_type = compression_config.get("algorithm")
-    return bool(algo_type is not None and algo_type == "quantization" and compression_config.get("params", {}))
+    if algo_type is not None and algo_type == "quantization" and compression_config.get("params", {}):
+        return True
+    return False
 
 
 def is_pretrained_model_requested(config: SampleConfig) -> bool:

nncf/common/pruning/operations.py

@@ -104,7 +104,9 @@ def mask_propagation(
 class ConvolutionPruningOp(BasePruningOp):
     @classmethod
     def accept_pruned_input(cls, node: NNCFNode) -> bool:
-        return not (is_grouped_conv(node) and not is_prunable_depthwise_conv(node))
+        if is_grouped_conv(node) and not is_prunable_depthwise_conv(node):
+            return False
+        return True
 
     @classmethod
     def mask_propagation(
@@ -124,7 +126,9 @@ def mask_propagation(
 class TransposeConvolutionPruningOp(BasePruningOp):
     @classmethod
     def accept_pruned_input(cls, node: NNCFNode) -> bool:
-        return not (is_grouped_conv(node) and not is_prunable_depthwise_conv(node))
+        if is_grouped_conv(node) and not is_prunable_depthwise_conv(node):
+            return False
+        return True
 
     @classmethod
     def mask_propagation(
@@ -274,7 +278,9 @@ def match_multiple_output_masks(
 
     @classmethod
     def accept_pruned_input(cls, node: NNCFNode):
-        return node.layer_attributes is not None
+        if node.layer_attributes is not None:
+            return True
+        return False
 
     @classmethod
     def generate_output_masks(
@@ -329,7 +335,9 @@ class PadPruningOp(IdentityMaskForwardPruningOp):
     @classmethod
     def accept_pruned_input(cls, node: NNCFNode) -> bool:
         mode, value = node.layer_attributes.mode, node.layer_attributes.value
-        return not (mode == "constant" and value != 0)
+        if mode == "constant" and value != 0:
+            return False
+        return True
 
 
 class ElementwisePruningOp(BasePruningOp):
@@ -386,7 +394,9 @@ def mask_propagation(
 class FlattenPruningOp(BasePruningOp):
     @classmethod
     def accept_pruned_input(cls, node: NNCFNode) -> bool:
-        return node.layer_attributes is not None
+        if node.layer_attributes is not None:
+            return True
+        return False
 
     @classmethod
     def mask_propagation(cls, node: NNCFNode, graph: NNCFGraph, tensor_processor: Type[NNCFPruningBaseTensorProcessor]):

nncf/onnx/graph/nncf_graph_builder.py

@@ -148,7 +148,7 @@ def _is_node_with_bias(node: onnx.NodeProto, model: onnx.ModelProto) -> bool:
     """
     metatype = get_metatype(model, node)
     bias_tensor_port_id = get_bias_tensor_port_id(metatype)
-    return bool(bias_tensor_port_id is not None and len(node.input) > bias_tensor_port_id)
+    return bias_tensor_port_id is not None and len(node.input) > bias_tensor_port_id
 
 
 def _get_gemm_attrs(node: onnx.NodeProto) -> Dict[str, int]:

nncf/quantization/algorithms/channel_alignment/algorithm.py

@@ -274,7 +274,9 @@ def _check_consumer_conv_node(self, conv_node: NNCFNode) -> bool:
         if any(elem != 1 for elem in attrs.stride):
             return False
         # Check Node has valid dilation
-        return not any(elem != 1 for elem in attrs.dilations)
+        if any(elem != 1 for elem in attrs.dilations):
+            return False
+        return True
 
     def _check_producer_conv_node(self, conv_node: NNCFNode):
         return conv_node.layer_attributes is not None

nncf/tensorflow/graph/transformations/commands.py

@@ -500,7 +500,7 @@ def commands(self) -> List[TFTransformationCommand]:
         return self._commands
 
     def check_insertion_command(self, command: TFTransformationCommand) -> bool:
-        return bool(
+        return (
             isinstance(command, TFTransformationCommand)
             and command.type == TransformationType.INSERT
             and self.check_target_points_fn(self.target_point, command.target_point)

nncf/torch/accuracy_aware_training/utils.py

@@ -15,7 +15,7 @@
 def is_dist_avail_and_initialized():
     if not dist.is_available():
         return False
-    return bool(dist.is_initialized())
+    return dist.is_initialized()
 
 
 def get_rank():

nncf/torch/graph/operator_metatypes.py

@@ -129,7 +129,7 @@ def matches(
     ) -> bool:
         if not isinstance(layer_attributes, ConvolutionLayerAttributes):
             return False
-        return bool(layer_attributes.groups == layer_attributes.in_channels and layer_attributes.in_channels > 1)
+        return layer_attributes.groups == layer_attributes.in_channels and layer_attributes.in_channels > 1
 
 
 @PT_OPERATOR_METATYPES.register()

nncf/torch/knowledge_distillation/knowledge_distillation_loss.py

@@ -141,7 +141,7 @@ def match_fn(obj):
     def _is_loss(obj):
         if not isinstance(obj, torch.Tensor):
             return False
-        return bool(obj.requires_grad)
+        return obj.requires_grad
 
     def forward(self) -> torch.Tensor:
         """

nncf/torch/utils.py

@@ -170,7 +170,7 @@ def get_flat_tensor_contents_string(input_tensor):
 def is_dist_avail_and_initialized():
     if not dist.is_available():
         return False
-    return bool(dist.is_initialized())
+    return dist.is_initialized()
 
 
 def get_rank():