[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Author: pre-commit-ci[bot]

lerobot/common/envs/factory.py

@@ -14,7 +14,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import importlib
-from collections import deque
 
 import gymnasium as gym
 

lerobot/common/optim/optimizers.py

@@ -99,52 +99,55 @@ def build(self, params: dict) -> torch.optim.Optimizer:
 @dataclass
 class MultiAdamConfig(OptimizerConfig):
     """Configuration for multiple Adam optimizers with different parameter groups.
-    
+
     This creates a dictionary of Adam optimizers, each with its own hyperparameters.
-    
+
     Args:
         lr: Default learning rate (used if not specified for a group)
         weight_decay: Default weight decay (used if not specified for a group)
         optimizer_groups: Dictionary mapping parameter group names to their hyperparameters
         grad_clip_norm: Gradient clipping norm
     """
+
     lr: float = 1e-3
     weight_decay: float = 0.0
     grad_clip_norm: float = 10.0
     optimizer_groups: dict[str, dict[str, Any]] = field(default_factory=dict)
-    
+
     def build(self, params_dict: dict[str, list]) -> dict[str, torch.optim.Optimizer]:
         """Build multiple Adam optimizers.
-        
+
         Args:
             params_dict: Dictionary mapping parameter group names to lists of parameters
                          The keys should match the keys in optimizer_groups
-        
+
         Returns:
             Dictionary mapping parameter group names to their optimizers
         """
         optimizers = {}
-        
+
         for name, params in params_dict.items():
             # Get group-specific hyperparameters or use defaults
             group_config = self.optimizer_groups.get(name, {})
-            
+
             # Create optimizer with merged parameters (defaults + group-specific)
             optimizer_kwargs = {
                 "lr": group_config.get("lr", self.lr),
                 "betas": group_config.get("betas", (0.9, 0.999)),
                 "eps": group_config.get("eps", 1e-5),
                 "weight_decay": group_config.get("weight_decay", self.weight_decay),
             }
-            
+
             optimizers[name] = torch.optim.Adam(params, **optimizer_kwargs)
-            
+
         return optimizers
 
 
-def save_optimizer_state(optimizer: torch.optim.Optimizer | dict[str, torch.optim.Optimizer], save_dir: Path) -> None:
+def save_optimizer_state(
+    optimizer: torch.optim.Optimizer | dict[str, torch.optim.Optimizer], save_dir: Path
+) -> None:
     """Save optimizer state to disk.
-    
+
     Args:
         optimizer: Either a single optimizer or a dictionary of optimizers.
         save_dir: Directory to save the optimizer state.
@@ -173,11 +176,11 @@ def load_optimizer_state(
     optimizer: torch.optim.Optimizer | dict[str, torch.optim.Optimizer], save_dir: Path
 ) -> torch.optim.Optimizer | dict[str, torch.optim.Optimizer]:
     """Load optimizer state from disk.
-    
+
     Args:
         optimizer: Either a single optimizer or a dictionary of optimizers.
         save_dir: Directory to load the optimizer state from.
-        
+
     Returns:
         The updated optimizer(s) with loaded state.
     """
@@ -201,7 +204,7 @@ def _load_single_optimizer_state(optimizer: torch.optim.Optimizer, save_dir: Pat
     current_state_dict = optimizer.state_dict()
     flat_state = load_file(save_dir / OPTIMIZER_STATE)
     state = unflatten_dict(flat_state)
-    
+
     # Handle case where 'state' key might not exist (for newly created optimizers)
     if "state" in state:
         loaded_state_dict = {"state": {int(k): v for k, v in state["state"].items()}}

lerobot/common/policies/factory.py

@@ -24,11 +24,11 @@
 from lerobot.common.envs.utils import env_to_policy_features
 from lerobot.common.policies.act.configuration_act import ACTConfig
 from lerobot.common.policies.diffusion.configuration_diffusion import DiffusionConfig
+from lerobot.common.policies.hilserl.classifier.configuration_classifier import ClassifierConfig
 from lerobot.common.policies.pi0.configuration_pi0 import PI0Config
 from lerobot.common.policies.pretrained import PreTrainedPolicy
 from lerobot.common.policies.tdmpc.configuration_tdmpc import TDMPCConfig
 from lerobot.common.policies.vqbet.configuration_vqbet import VQBeTConfig
-from lerobot.common.policies.hilserl.classifier.configuration_classifier import ClassifierConfig
 from lerobot.configs.policies import PreTrainedConfig
 from lerobot.configs.types import FeatureType
 

lerobot/common/policies/hilserl/classifier/configuration_classifier.py

@@ -1,10 +1,9 @@
-from dataclasses import dataclass, field
-from typing import Dict, List
+from dataclasses import dataclass
+from typing import List
 
 from lerobot.common.optim.optimizers import AdamWConfig, OptimizerConfig
 from lerobot.common.optim.schedulers import LRSchedulerConfig
 from lerobot.configs.policies import PreTrainedConfig
-from lerobot.configs.types import FeatureType, PolicyFeature
 
 
 @PreTrainedConfig.register_subclass(name="hilserl_classifier")

lerobot/common/policies/normalize.py

@@ -82,8 +82,10 @@ def create_stats_buffers(
         if stats and key in stats:
             if norm_mode is NormalizationMode.MEAN_STD:
                 if "mean" not in stats[key] or "std" not in stats[key]:
-                    raise ValueError(f"Missing 'mean' or 'std' in stats for key {key} with MEAN_STD normalization")
-                
+                    raise ValueError(
+                        f"Missing 'mean' or 'std' in stats for key {key} with MEAN_STD normalization"
+                    )
+
                 if isinstance(stats[key]["mean"], np.ndarray):
                     buffer["mean"].data = torch.from_numpy(stats[key]["mean"]).to(dtype=torch.float32)
                     buffer["std"].data = torch.from_numpy(stats[key]["std"]).to(dtype=torch.float32)
@@ -96,12 +98,16 @@ def create_stats_buffers(
                     buffer["std"].data = stats[key]["std"].clone().to(dtype=torch.float32)
                 else:
                     type_ = type(stats[key]["mean"])
-                    raise ValueError(f"np.ndarray or torch.Tensor expected for 'mean', but type is '{type_}' instead.")
-                    
+                    raise ValueError(
+                        f"np.ndarray or torch.Tensor expected for 'mean', but type is '{type_}' instead."
+                    )
+
             elif norm_mode is NormalizationMode.MIN_MAX:
                 if "min" not in stats[key] or "max" not in stats[key]:
-                    raise ValueError(f"Missing 'min' or 'max' in stats for key {key} with MIN_MAX normalization")
-                
+                    raise ValueError(
+                        f"Missing 'min' or 'max' in stats for key {key} with MIN_MAX normalization"
+                    )
+
                 if isinstance(stats[key]["min"], np.ndarray):
                     buffer["min"].data = torch.from_numpy(stats[key]["min"]).to(dtype=torch.float32)
                     buffer["max"].data = torch.from_numpy(stats[key]["max"]).to(dtype=torch.float32)
@@ -110,7 +116,9 @@ def create_stats_buffers(
                     buffer["max"].data = stats[key]["max"].clone().to(dtype=torch.float32)
                 else:
                     type_ = type(stats[key]["min"])
-                    raise ValueError(f"np.ndarray or torch.Tensor expected for 'min', but type is '{type_}' instead.")
+                    raise ValueError(
+                        f"np.ndarray or torch.Tensor expected for 'min', but type is '{type_}' instead."
+                    )
 
         stats_buffers[key] = buffer
     return stats_buffers

lerobot/common/policies/sac/configuration_sac.py

@@ -19,7 +19,7 @@
 
 from lerobot.common.optim.optimizers import MultiAdamConfig
 from lerobot.configs.policies import PreTrainedConfig
-from lerobot.configs.types import FeatureType, NormalizationMode, PolicyFeature
+from lerobot.configs.types import NormalizationMode
 
 
 @dataclass

lerobot/common/policies/sac/modeling_sac.py

@@ -897,7 +897,6 @@ def _convert_normalization_params_to_tensor(normalization_params: dict) -> dict:
     #     for j in range(i + 1, num_critics):
     #         diff = torch.abs(q_values[i] - q_values[j]).mean().item()
     #         print(f"Mean difference between critic {i} and {j}: {diff:.6f}")
-    import draccus
 
     from lerobot.configs import parser
 

lerobot/common/utils/wandb_utils.py

@@ -115,11 +115,13 @@ def log_policy(self, checkpoint_dir: Path):
         artifact.add_file(checkpoint_dir / PRETRAINED_MODEL_DIR / SAFETENSORS_SINGLE_FILE)
         self._wandb.log_artifact(artifact)
 
-    def log_dict(self, d: dict, step: int | None = None, mode: str = "train", custom_step_key: str | None = None):
+    def log_dict(
+        self, d: dict, step: int | None = None, mode: str = "train", custom_step_key: str | None = None
+    ):
         if mode not in {"train", "eval"}:
             raise ValueError(mode)
         if step is None and custom_step_key is None:
-                    raise ValueError("Either step or custom_step_key must be provided.")
+            raise ValueError("Either step or custom_step_key must be provided.")
 
         # NOTE: This is not simple. Wandb step is it must always monotonically increase and it
         # increases with each wandb.log call, but in the case of asynchronous RL for example,
@@ -142,10 +144,7 @@ def log_dict(self, d: dict, step: int | None = None, mode: str = "train", custom
                 continue
 
             # Do not log the custom step key itself.
-            if (
-                self._wandb_custom_step_key is not None
-                and k in self._wandb_custom_step_key
-            ):
+            if self._wandb_custom_step_key is not None and k in self._wandb_custom_step_key:
                 continue
 
             if custom_step_key is not None:
@@ -160,7 +159,6 @@ def log_dict(self, d: dict, step: int | None = None, mode: str = "train", custom
 
             self._wandb.log(data={f"{mode}/{k}": v}, step=step)
 
-
     def log_video(self, video_path: str, step: int, mode: str = "train"):
         if mode not in {"train", "eval"}:
             raise ValueError(mode)

lerobot/configs/train.py

@@ -34,7 +34,7 @@
 
 @dataclass
 class TrainPipelineConfig(HubMixin):
-    dataset: DatasetConfig | None = None # NOTE: In RL, we don't need a dataset
+    dataset: DatasetConfig | None = None  # NOTE: In RL, we don't need a dataset
     env: envs.EnvConfig | None = None
     policy: PreTrainedConfig | None = None
     # Set `dir` to where you would like to save all of the run outputs. If you run another training session # with the same value for `dir` its contents will be overwritten unless you set `resume` to true.

lerobot/scripts/server/actor_server.py

@@ -47,7 +47,7 @@
     python_object_to_bytes,
     transitions_to_bytes,
 )
-from lerobot.scripts.server.gym_manipulator import get_classifier, make_robot_env
+from lerobot.scripts.server.gym_manipulator import make_robot_env
 from lerobot.scripts.server.network_utils import (
     receive_bytes_in_chunks,
     send_bytes_in_chunks,
@@ -444,7 +444,7 @@ def receive_policy(
 
         # Initialize logging with explicit log file
         init_logging(log_file=log_file)
-        logging.info(f"Actor receive policy process logging initialized")
+        logging.info("Actor receive policy process logging initialized")
 
         # Setup process handlers to handle shutdown signal
         # But use shutdown event from the main process