[HIL-SERL] Review feedback modifications (#1112)

lerobot/common/envs/factory.py

@@ -68,6 +68,7 @@ def make_env(cfg: EnvConfig, n_envs: int = 1, use_async_envs: bool = False) -> g
         [lambda: gym.make(gym_handle, disable_env_checker=True, **cfg.gym_kwargs) for _ in range(n_envs)]
     )
     # TODO: add observation processor wrapper and remove preprocess_observation in the codebase
+    # https://github.com/Farama-Foundation/Gymnasium/blob/main/gymnasium/wrappers/vector/vectorize_observation.py#L19,
     # env = ObservationProcessorWrapper(env=env)
 
     return env

lerobot/common/envs/utils.py

@@ -81,35 +81,6 @@ def preprocess_observation(observations: dict[str, np.ndarray]) -> dict[str, Ten
     return return_observations
 
 
-class ObservationProcessorWrapper(gym.vector.VectorEnvWrapper):
-    def __init__(self, env: gym.vector.VectorEnv):
-        super().__init__(env)
-
-    def _observations(self, observations: dict[str, Any]) -> dict[str, Any]:
-        return preprocess_observation(observations)
-
-    def reset(
-        self,
-        *,
-        seed: int | list[int] | None = None,
-        options: dict[str, Any] | None = None,
-    ):
-        """Modifies the observation returned from the environment ``reset`` using the :meth:`observation`."""
-        observations, infos = self.env.reset(seed=seed, options=options)
-        return self._observations(observations), infos
-
-    def step(self, actions):
-        """Modifies the observation returned from the environment ``step`` using the :meth:`observation`."""
-        observations, rewards, terminations, truncations, infos = self.env.step(actions)
-        return (
-            self._observations(observations),
-            rewards,
-            terminations,
-            truncations,
-            infos,
-        )
-
-
 def env_to_policy_features(env_cfg: EnvConfig) -> dict[str, PolicyFeature]:
     # TODO(aliberts, rcadene): remove this hardcoding of keys and just use the nested keys as is
     # (need to also refactor preprocess_observation and externalize normalization from policies)

lerobot/common/optim/optimizers.py

@@ -46,6 +46,15 @@ class OptimizerConfig(draccus.ChoiceRegistry, abc.ABC):
 
     @abc.abstractmethod
     def build(self) -> torch.optim.Optimizer | dict[str, torch.optim.Optimizer]:
+        """
+        Build the optimizer. It can be a single optimizer or a dictionary of optimizers.
+        NOTE: Multiple optimizers are useful when you have different models to optimize.
+        For example, you can have one optimizer for the policy and another one for the value function
+        in reinforcement learning settings.
+
+        Returns:
+            The optimizer or a dictionary of optimizers.
+        """
         raise NotImplementedError
 
 

lerobot/common/policies/normalize.py

@@ -79,48 +79,28 @@ def create_stats_buffers(
             )
 
         # TODO(aliberts, rcadene): harmonize this to only use one framework (np or torch)
-        if stats and key in stats:
+        if stats:
-            # NOTE:(maractingi, azouitine): Change the order of these conditions because in online environments we don't have dataset stats
+            if isinstance(stats[key]["mean"], np.ndarray):
-            # Therefore, we don't access to full stats of the data, some elements either have min-max or mean-std only
+                if norm_mode is NormalizationMode.MEAN_STD:
-            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"
-                    )
-
-                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)
-                elif isinstance(stats[key]["mean"], torch.Tensor):
+                elif norm_mode is NormalizationMode.MIN_MAX:
-                    # Note: The clone is needed to make sure that the logic in save_pretrained doesn't see duplicated
-                    # tensors anywhere (for example, when we use the same stats for normalization and
-                    # unnormalization). See the logic here
-                    # https://github.com/huggingface/safetensors/blob/079781fd0dc455ba0fe851e2b4507c33d0c0d407/bindings/python/py_src/safetensors/torch.py#L97.
-                    buffer["mean"].data = stats[key]["mean"].clone().to(dtype=torch.float32)
-                    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."
-                    )
-
-            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"
-                    )
-
-                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)
-                elif isinstance(stats[key]["min"], torch.Tensor):
+            elif isinstance(stats[key]["mean"], torch.Tensor):
+                # Note: The clone is needed to make sure that the logic in save_pretrained doesn't see duplicated
+                # tensors anywhere (for example, when we use the same stats for normalization and
+                # unnormalization). See the logic here
+                # https://github.com/huggingface/safetensors/blob/079781fd0dc455ba0fe851e2b4507c33d0c0d407/bindings/python/py_src/safetensors/torch.py#L97.
+                if norm_mode is NormalizationMode.MEAN_STD:
+                    buffer["mean"].data = stats[key]["mean"].clone().to(dtype=torch.float32)
+                    buffer["std"].data = stats[key]["std"].clone().to(dtype=torch.float32)
+                elif norm_mode is NormalizationMode.MIN_MAX:
                     buffer["min"].data = stats[key]["min"].clone().to(dtype=torch.float32)
                     buffer["max"].data = stats[key]["max"].clone().to(dtype=torch.float32)
-                else:
+            else:
-                    type_ = type(stats[key]["min"])
+                type_ = type(stats[key]["mean"])
-                    raise ValueError(
+                raise ValueError(f"np.ndarray or torch.Tensor expected, but type is '{type_}' instead.")
-                        f"np.ndarray or torch.Tensor expected for 'min', but type is '{type_}' instead."
-                    )
 
         stats_buffers[key] = buffer
     return stats_buffers
@@ -169,13 +149,12 @@ class Normalize(nn.Module):
             setattr(self, "buffer_" + key.replace(".", "_"), buffer)
 
     # TODO(rcadene): should we remove torch.no_grad?
-    # @torch.no_grad
+    @torch.no_grad
     def forward(self, batch: dict[str, Tensor]) -> dict[str, Tensor]:
         batch = dict(batch)  # shallow copy avoids mutating the input batch
         for key, ft in self.features.items():
             if key not in batch:
                 # FIXME(aliberts, rcadene): This might lead to silent fail!
-                # NOTE: (azouitine) This continues help us for instantiation SACPolicy
                 continue
 
             norm_mode = self.norm_map.get(ft.type, NormalizationMode.IDENTITY)
@@ -244,7 +223,7 @@ class Unnormalize(nn.Module):
             setattr(self, "buffer_" + key.replace(".", "_"), buffer)
 
     # TODO(rcadene): should we remove torch.no_grad?
-    # @torch.no_grad
+    @torch.no_grad
     def forward(self, batch: dict[str, Tensor]) -> dict[str, Tensor]:
         batch = dict(batch)  # shallow copy avoids mutating the input batch
         for key, ft in self.features.items():
@@ -273,3 +252,170 @@ class Unnormalize(nn.Module):
             else:
                 raise ValueError(norm_mode)
         return batch
+
+
+# TODO: We should replace all normalization on the policies with register_buffer normalization
+#       and remove the `Normalize` and `Unnormalize` classes.
+def _initialize_stats_buffers(
+    module: nn.Module,
+    features: dict[str, PolicyFeature],
+    norm_map: dict[str, NormalizationMode],
+    stats: dict[str, dict[str, Tensor]] | None = None,
+) -> None:
+    """Register statistics buffers (mean/std or min/max) on the given *module*.
+
+    The logic matches the previous constructors of `NormalizeBuffer` and `UnnormalizeBuffer`,
+    but is factored out so it can be reused by both classes and stay in sync.
+    """
+    for key, ft in features.items():
+        norm_mode = norm_map.get(ft.type, NormalizationMode.IDENTITY)
+        if norm_mode is NormalizationMode.IDENTITY:
+            continue
+
+        shape: tuple[int, ...] = tuple(ft.shape)
+        if ft.type is FeatureType.VISUAL:
+            # reduce spatial dimensions, keep channel dimension only
+            c, *_ = shape
+            shape = (c, 1, 1)
+
+        prefix = key.replace(".", "_")
+
+        if norm_mode is NormalizationMode.MEAN_STD:
+            mean = torch.full(shape, torch.inf, dtype=torch.float32)
+            std = torch.full(shape, torch.inf, dtype=torch.float32)
+
+            if stats and key in stats and "mean" in stats[key] and "std" in stats[key]:
+                mean_data = stats[key]["mean"]
+                std_data = stats[key]["std"]
+                if isinstance(mean_data, np.ndarray):
+                    mean = torch.from_numpy(mean_data).to(dtype=torch.float32)
+                    std = torch.from_numpy(std_data).to(dtype=torch.float32)
+                elif isinstance(mean_data, torch.Tensor):
+                    mean = mean_data.clone().to(dtype=torch.float32)
+                    std = std_data.clone().to(dtype=torch.float32)
+                else:
+                    raise ValueError(f"Unsupported stats type for key '{key}' (expected ndarray or Tensor).")
+
+            module.register_buffer(f"{prefix}_mean", mean)
+            module.register_buffer(f"{prefix}_std", std)
+            continue
+
+        if norm_mode is NormalizationMode.MIN_MAX:
+            min_val = torch.full(shape, torch.inf, dtype=torch.float32)
+            max_val = torch.full(shape, torch.inf, dtype=torch.float32)
+
+            if stats and key in stats and "min" in stats[key] and "max" in stats[key]:
+                min_data = stats[key]["min"]
+                max_data = stats[key]["max"]
+                if isinstance(min_data, np.ndarray):
+                    min_val = torch.from_numpy(min_data).to(dtype=torch.float32)
+                    max_val = torch.from_numpy(max_data).to(dtype=torch.float32)
+                elif isinstance(min_data, torch.Tensor):
+                    min_val = min_data.clone().to(dtype=torch.float32)
+                    max_val = max_data.clone().to(dtype=torch.float32)
+                else:
+                    raise ValueError(f"Unsupported stats type for key '{key}' (expected ndarray or Tensor).")
+
+            module.register_buffer(f"{prefix}_min", min_val)
+            module.register_buffer(f"{prefix}_max", max_val)
+            continue
+
+        raise ValueError(norm_mode)
+
+
+class NormalizeBuffer(nn.Module):
+    """Same as `Normalize` but statistics are stored as registered buffers rather than parameters."""
+
+    def __init__(
+        self,
+        features: dict[str, PolicyFeature],
+        norm_map: dict[str, NormalizationMode],
+        stats: dict[str, dict[str, Tensor]] | None = None,
+    ):
+        super().__init__()
+        self.features = features
+        self.norm_map = norm_map
+
+        _initialize_stats_buffers(self, features, norm_map, stats)
+
+    def forward(self, batch: dict[str, Tensor]) -> dict[str, Tensor]:
+        batch = dict(batch)
+        for key, ft in self.features.items():
+            if key not in batch:
+                continue
+
+            norm_mode = self.norm_map.get(ft.type, NormalizationMode.IDENTITY)
+            if norm_mode is NormalizationMode.IDENTITY:
+                continue
+
+            prefix = key.replace(".", "_")
+
+            if norm_mode is NormalizationMode.MEAN_STD:
+                mean = getattr(self, f"{prefix}_mean")
+                std = getattr(self, f"{prefix}_std")
+                assert not torch.isinf(mean).any(), _no_stats_error_str("mean")
+                assert not torch.isinf(std).any(), _no_stats_error_str("std")
+                batch[key] = (batch[key] - mean) / (std + 1e-8)
+                continue
+
+            if norm_mode is NormalizationMode.MIN_MAX:
+                min_val = getattr(self, f"{prefix}_min")
+                max_val = getattr(self, f"{prefix}_max")
+                assert not torch.isinf(min_val).any(), _no_stats_error_str("min")
+                assert not torch.isinf(max_val).any(), _no_stats_error_str("max")
+                batch[key] = (batch[key] - min_val) / (max_val - min_val + 1e-8)
+                batch[key] = batch[key] * 2 - 1
+                continue
+
+            raise ValueError(norm_mode)
+
+        return batch
+
+
+class UnnormalizeBuffer(nn.Module):
+    """Inverse operation of `NormalizeBuffer`. Uses registered buffers for statistics."""
+
+    def __init__(
+        self,
+        features: dict[str, PolicyFeature],
+        norm_map: dict[str, NormalizationMode],
+        stats: dict[str, dict[str, Tensor]] | None = None,
+    ):
+        super().__init__()
+        self.features = features
+        self.norm_map = norm_map
+
+        _initialize_stats_buffers(self, features, norm_map, stats)
+
+    def forward(self, batch: dict[str, Tensor]) -> dict[str, Tensor]:
+        batch = dict(batch)
+        for key, ft in self.features.items():
+            if key not in batch:
+                continue
+
+            norm_mode = self.norm_map.get(ft.type, NormalizationMode.IDENTITY)
+            if norm_mode is NormalizationMode.IDENTITY:
+                continue
+
+            prefix = key.replace(".", "_")
+
+            if norm_mode is NormalizationMode.MEAN_STD:
+                mean = getattr(self, f"{prefix}_mean")
+                std = getattr(self, f"{prefix}_std")
+                assert not torch.isinf(mean).any(), _no_stats_error_str("mean")
+                assert not torch.isinf(std).any(), _no_stats_error_str("std")
+                batch[key] = batch[key] * std + mean
+                continue
+
+            if norm_mode is NormalizationMode.MIN_MAX:
+                min_val = getattr(self, f"{prefix}_min")
+                max_val = getattr(self, f"{prefix}_max")
+                assert not torch.isinf(min_val).any(), _no_stats_error_str("min")
+                assert not torch.isinf(max_val).any(), _no_stats_error_str("max")
+                batch[key] = (batch[key] + 1) / 2
+                batch[key] = batch[key] * (max_val - min_val) + min_val
+                continue
+
+            raise ValueError(norm_mode)
+
+        return batch

lerobot/common/policies/reward_model/modeling_classifier.py

@@ -9,9 +9,6 @@ from lerobot.common.policies.normalize import Normalize, Unnormalize
 from lerobot.common.policies.pretrained import PreTrainedPolicy
 from lerobot.common.policies.reward_model.configuration_classifier import RewardClassifierConfig
 
-logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
-logger = logging.getLogger(__name__)
-
 
 class ClassifierOutput:
     """Wrapper for classifier outputs with additional metadata."""

lerobot/common/policies/sac/configuration_sac.py

@@ -24,6 +24,12 @@ from lerobot.configs.types import NormalizationMode
 
 @dataclass
 class ConcurrencyConfig:
+    """Configuration for the concurrency of the actor and learner.
+    Possible values are:
+    - "threads": Use threads for the actor and learner.
+    - "processes": Use processes for the actor and learner.
+    """
+
     actor: str = "threads"
     learner: str = "threads"
 
@@ -68,51 +74,9 @@ class SACConfig(PreTrainedConfig):
     This configuration class contains all the parameters needed to define a SAC agent,
     including network architectures, optimization settings, and algorithm-specific
     hyperparameters.
-
-    Args:
-        actor_network_kwargs: Configuration for the actor network architecture.
-        critic_network_kwargs: Configuration for the critic network architecture.
-        discrete_critic_network_kwargs: Configuration for the discrete critic network.
-        policy_kwargs: Configuration for the policy parameters.
-        n_obs_steps: Number of observation steps to consider.
-        normalization_mapping: Mapping of feature types to normalization modes.
-        dataset_stats: Statistics for normalizing different types of inputs.
-        input_features: Dictionary of input features with their types and shapes.
-        output_features: Dictionary of output features with their types and shapes.
-        camera_number: Number of cameras used for visual observations.
-        device: Device to run the model on (e.g., "cuda", "cpu").
-        storage_device: Device to store the model on.
-        vision_encoder_name: Name of the vision encoder model.
-        freeze_vision_encoder: Whether to freeze the vision encoder during training.
-        image_encoder_hidden_dim: Hidden dimension size for the image encoder.
-        shared_encoder: Whether to use a shared encoder for actor and critic.
-        num_discrete_actions: Number of discrete actions, eg for gripper actions.
-        image_embedding_pooling_dim: Dimension of the image embedding pooling.
-        concurrency: Configuration for concurrency settings.
-        actor_learner_config: Configuration for actor-learner architecture.
-        online_steps: Number of steps for online training.
-        online_env_seed: Seed for the online environment.
-        online_buffer_capacity: Capacity of the online replay buffer.
-        offline_buffer_capacity: Capacity of the offline replay buffer.
-        async_prefetch: Whether to use asynchronous prefetching for the buffers.
-        online_step_before_learning: Number of steps before learning starts.
-        policy_update_freq: Frequency of policy updates.
-        discount: Discount factor for the SAC algorithm.
-        temperature_init: Initial temperature value.
-        num_critics: Number of critics in the ensemble.
-        num_subsample_critics: Number of subsampled critics for training.
-        critic_lr: Learning rate for the critic network.
-        actor_lr: Learning rate for the actor network.
-        temperature_lr: Learning rate for the temperature parameter.
-        critic_target_update_weight: Weight for the critic target update.
-        utd_ratio: Update-to-data ratio for the UTD algorithm.
-        state_encoder_hidden_dim: Hidden dimension size for the state encoder.
-        latent_dim: Dimension of the latent space.
-        target_entropy: Target entropy for the SAC algorithm.
-        use_backup_entropy: Whether to use backup entropy for the SAC algorithm.
-        grad_clip_norm: Gradient clipping norm for the SAC algorithm.
     """
 
+    # Mapping of feature types to normalization modes
     normalization_mapping: dict[str, NormalizationMode] = field(
         default_factory=lambda: {
             "VISUAL": NormalizationMode.MEAN_STD,
@@ -122,6 +86,7 @@ class SACConfig(PreTrainedConfig):
         }
     )
 
+    # Statistics for normalizing different types of inputs
     dataset_stats: dict[str, dict[str, list[float]]] | None = field(
         default_factory=lambda: {
             "observation.image": {
@@ -140,47 +105,81 @@ class SACConfig(PreTrainedConfig):
     )
 
     # Architecture specifics
+    # Device to run the model on (e.g., "cuda", "cpu")
     device: str = "cpu"
+    # Device to store the model on
     storage_device: str = "cpu"
-    # Set to "helper2424/resnet10" for hil serl
+    # Name of the vision encoder model (Set to "helper2424/resnet10" for hil serl resnet10)
     vision_encoder_name: str | None = None
+    # Whether to freeze the vision encoder during training
     freeze_vision_encoder: bool = True
+    # Hidden dimension size for the image encoder
     image_encoder_hidden_dim: int = 32
+    # Whether to use a shared encoder for actor and critic
     shared_encoder: bool = True
+    # Number of discrete actions, eg for gripper actions
     num_discrete_actions: int | None = None
+    # Dimension of the image embedding pooling
     image_embedding_pooling_dim: int = 8
 
     # Training parameter
+    # Number of steps for online training
     online_steps: int = 1000000
+    # Seed for the online environment
     online_env_seed: int = 10000
+    # Capacity of the online replay buffer
     online_buffer_capacity: int = 100000
+    # Capacity of the offline replay buffer
     offline_buffer_capacity: int = 100000
+    # Whether to use asynchronous prefetching for the buffers
     async_prefetch: bool = False
+    # Number of steps before learning starts
     online_step_before_learning: int = 100
+    # Frequency of policy updates
     policy_update_freq: int = 1
 
     # SAC algorithm parameters
+    # Discount factor for the SAC algorithm
     discount: float = 0.99
+    # Initial temperature value
     temperature_init: float = 1.0
+    # Number of critics in the ensemble
     num_critics: int = 2
+    # Number of subsampled critics for training
     num_subsample_critics: int | None = None
+    # Learning rate for the critic network
     critic_lr: float = 3e-4
+    # Learning rate for the actor network
     actor_lr: float = 3e-4
+    # Learning rate for the temperature parameter
     temperature_lr: float = 3e-4
+    # Weight for the critic target update
     critic_target_update_weight: float = 0.005
-    utd_ratio: int = 1  # If you want enable utd_ratio, you need to set it to >1
+    # Update-to-data ratio for the UTD algorithm (If you want enable utd_ratio, you need to set it to >1)
+    utd_ratio: int = 1
+    # Hidden dimension size for the state encoder
     state_encoder_hidden_dim: int = 256
+    # Dimension of the latent space
     latent_dim: int = 256
+    # Target entropy for the SAC algorithm
     target_entropy: float | None = None
+    # Whether to use backup entropy for the SAC algorithm
     use_backup_entropy: bool = True
+    # Gradient clipping norm for the SAC algorithm
     grad_clip_norm: float = 40.0
 
     # Network configuration
+    # Configuration for the critic network architecture
     critic_network_kwargs: CriticNetworkConfig = field(default_factory=CriticNetworkConfig)
+    # Configuration for the actor network architecture
     actor_network_kwargs: ActorNetworkConfig = field(default_factory=ActorNetworkConfig)
+    # Configuration for the policy parameters
     policy_kwargs: PolicyConfig = field(default_factory=PolicyConfig)
+    # Configuration for the discrete critic network
     discrete_critic_network_kwargs: CriticNetworkConfig = field(default_factory=CriticNetworkConfig)
+    # Configuration for actor-learner architecture
     actor_learner_config: ActorLearnerConfig = field(default_factory=ActorLearnerConfig)
+    # Configuration for concurrency settings (you can use threads or processes for the actor and learner)
     concurrency: ConcurrencyConfig = field(default_factory=ConcurrencyConfig)
 
     # Optimizations

lerobot/common/policies/sac/modeling_sac.py

@@ -27,7 +27,7 @@ import torch.nn.functional as F  # noqa: N812
 from torch import Tensor
 from torch.distributions import MultivariateNormal, TanhTransform, Transform, TransformedDistribution
 
-from lerobot.common.policies.normalize import Normalize, Unnormalize
+from lerobot.common.policies.normalize import NormalizeBuffer, UnnormalizeBuffer
 from lerobot.common.policies.pretrained import PreTrainedPolicy
 from lerobot.common.policies.sac.configuration_sac import SACConfig
 from lerobot.common.policies.utils import get_device_from_parameters
@@ -394,17 +394,16 @@ class SACPolicy(
         self.normalize_inputs = nn.Identity()
         self.normalize_targets = nn.Identity()
         self.unnormalize_outputs = nn.Identity()
-
+        if self.config.dataset_stats is not None:
-        if self.config.dataset_stats:
             params = _convert_normalization_params_to_tensor(self.config.dataset_stats)
-            self.normalize_inputs = Normalize(
+            self.normalize_inputs = NormalizeBuffer(
                 self.config.input_features, self.config.normalization_mapping, params
             )
             stats = dataset_stats or params
-            self.normalize_targets = Normalize(
+            self.normalize_targets = NormalizeBuffer(
                 self.config.output_features, self.config.normalization_mapping, stats
             )
-            self.unnormalize_outputs = Unnormalize(
+            self.unnormalize_outputs = UnnormalizeBuffer(
                 self.config.output_features, self.config.normalization_mapping, stats
             )
 
@@ -506,7 +505,7 @@ class SACObservationEncoder(nn.Module):
         if not self.has_images:
             return
 
-        if self.config.vision_encoder_name:
+        if self.config.vision_encoder_name is not None:
             self.image_encoder = PretrainedImageEncoder(self.config)
         else:
             self.image_encoder = DefaultImageEncoder(self.config)

lerobot/common/utils/utils.py

@@ -19,9 +19,10 @@ import os.path as osp
 import platform
 import subprocess
 import time
-from copy import copy
+from copy import copy, deepcopy
 from datetime import datetime, timezone
 from pathlib import Path
+from statistics import mean
 
 import numpy as np
 import torch
@@ -108,11 +109,14 @@ def is_amp_available(device: str):
         raise ValueError(f"Unknown device '{device}.")
 
 
-def init_logging(log_file=None):
+def init_logging(log_file: Path | None = None, display_pid: bool = False):
     def custom_format(record):
         dt = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
         fnameline = f"{record.pathname}:{record.lineno}"
-        message = f"{record.levelname} [PID: {os.getpid()}] {dt} {fnameline[-15:]:>15} {record.msg}"
+
+        # NOTE: Display PID is useful for multi-process logging.
+        pid_str = f"[PID: {os.getpid()}]" if display_pid else ""
+        message = f"{record.levelname} {pid_str} {dt} {fnameline[-15:]:>15} {record.msg}"
         return message
 
     logging.basicConfig(level=logging.INFO)
@@ -238,30 +242,99 @@ def is_valid_numpy_dtype_string(dtype_str: str) -> bool:
 
 
 class TimerManager:
+    """
+    Lightweight utility to measure elapsed time.
+
+    Examples
+    --------
+    >>> timer = TimerManager("Policy", log=False)
+    >>> for _ in range(3):
+    ...     with timer:
+    ...         time.sleep(0.01)
+    >>> print(timer.last, timer.fps_avg, timer.percentile(90))
+    """
+
     def __init__(
         self,
-        elapsed_time_list: list[float] | None = None,
+        label: str = "Elapsed-time",
-        label="Elapsed time",
+        log: bool = True,
-        log=True,
+        logger: logging.Logger | None = None,
     ):
         self.label = label
-        self.elapsed_time_list = elapsed_time_list
         self.log = log
-        self.elapsed = 0.0
+        self.logger = logger
+        self._start: float | None = None
+        self._history: list[float] = []
 
     def __enter__(self):
-        self.start = time.perf_counter()
+        return self.start()
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        self.stop()
+
+    def start(self):
+        self._start = time.perf_counter()
         return self
 
-    def __exit__(self, exc_type, exc_value, traceback):
+    def stop(self) -> float:
-        self.elapsed: float = time.perf_counter() - self.start
+        if self._start is None:
-
+            raise RuntimeError("Timer was never started.")
-        if self.elapsed_time_list is not None:
+        elapsed = time.perf_counter() - self._start
-            self.elapsed_time_list.append(self.elapsed)
+        self._history.append(elapsed)
-
+        self._start = None
         if self.log:
-            print(f"{self.label}: {self.elapsed:.6f} seconds")
+            if self.logger is not None:
+                self.logger.info(f"{self.label}: {elapsed:.6f} s")
+            else:
+                logging.info(f"{self.label}: {elapsed:.6f} s")
+        return elapsed
+
+    def reset(self):
+        self._history.clear()
 
     @property
-    def elapsed_seconds(self):
+    def last(self) -> float:
-        return self.elapsed
+        return self._history[-1] if self._history else 0.0
+
+    @property
+    def avg(self) -> float:
+        return mean(self._history) if self._history else 0.0
+
+    @property
+    def total(self) -> float:
+        return sum(self._history)
+
+    @property
+    def count(self) -> int:
+        return len(self._history)
+
+    @property
+    def history(self) -> list[float]:
+        return deepcopy(self._history)
+
+    @property
+    def fps_history(self) -> list[float]:
+        return [1.0 / t for t in self._history]
+
+    @property
+    def fps_last(self) -> float:
+        return 0.0 if self.last == 0 else 1.0 / self.last
+
+    @property
+    def fps_avg(self) -> float:
+        return 0.0 if self.avg == 0 else 1.0 / self.avg
+
+    def percentile(self, p: float) -> float:
+        """
+        Return the p-th percentile of recorded times.
+        """
+        if not self._history:
+            return 0.0
+        return float(np.percentile(self._history, p))
+
+    def fps_percentile(self, p: float) -> float:
+        """
+        FPS corresponding to the p-th percentile time.
+        """
+        val = self.percentile(p)
+        return 0.0 if val == 0 else 1.0 / val

lerobot/common/utils/wandb_utils.py

@@ -123,9 +123,9 @@ class WandBLogger:
         if step is None and custom_step_key is None:
             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
+        # NOTE: This is not simple. Wandb step must always monotonically increase and it
         # increases with each wandb.log call, but in the case of asynchronous RL for example,
-        # multiple time steps is possible for example, the interaction step with the environment,
+        # multiple time steps is possible. For example, the interaction step with the environment,
         # the training step, the evaluation step, etc. So we need to define a custom step key
         # to log the correct step for each metric.
         if custom_step_key is not None:

lerobot/scripts/server/actor_server.py

@@ -13,13 +13,67 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+"""
+Actor server runner for distributed HILSerl robot policy training.
+
+This script implements the actor component of the distributed HILSerl architecture.
+It executes the policy in the robot environment, collects experience,
+and sends transitions to the learner server for policy updates.
+
+Examples of usage:
+
+- Start an actor server for real robot training with human-in-the-loop intervention:
+```bash
+python lerobot/scripts/server/actor_server.py --config_path lerobot/configs/train_config_hilserl_so100.json
+```
+
+- Run with a specific robot type for a pick and place task:
+```bash
+python lerobot/scripts/server/actor_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --robot.type=so100 \
+    --task=pick_and_place
+```
+
+- Set a custom workspace bound for the robot's end-effector:
+```bash
+python lerobot/scripts/server/actor_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --env.ee_action_space_params.bounds.max="[0.24, 0.20, 0.10]" \
+    --env.ee_action_space_params.bounds.min="[0.16, -0.08, 0.03]"
+```
+
+- Run with specific camera crop parameters:
+```bash
+python lerobot/scripts/server/actor_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --env.crop_params_dict="{'observation.images.side': [180, 207, 180, 200], 'observation.images.front': [180, 250, 120, 150]}"
+```
+
+**NOTE**: The actor server requires a running learner server to connect to. Ensure the learner
+server is started before launching the actor.
+
+**NOTE**: Human intervention is key to HILSerl training. Press the upper right trigger button on the
+gamepad to take control of the robot during training. Initially intervene frequently, then gradually
+reduce interventions as the policy improves.
+
+**WORKFLOW**:
+1. Determine robot workspace bounds using `find_joint_limits.py`
+2. Record demonstrations with `gym_manipulator.py` in record mode
+3. Process the dataset and determine camera crops with `crop_dataset_roi.py`
+4. Start the learner server with the training configuration
+5. Start this actor server with the same configuration
+6. Use human interventions to guide policy learning
+
+For more details on the complete HILSerl training workflow, see:
+https://github.com/michel-aractingi/lerobot-hilserl-guide
+"""
 
 import logging
 import os
 import time
 from functools import lru_cache
 from queue import Empty
-from statistics import mean, quantiles
 
 import grpc
 import torch
@@ -65,10 +119,12 @@ ACTOR_SHUTDOWN_TIMEOUT = 30
 @parser.wrap()
 def actor_cli(cfg: TrainPipelineConfig):
     cfg.validate()
+    display_pid = False
     if not use_threads(cfg):
         import torch.multiprocessing as mp
 
         mp.set_start_method("spawn")
+        display_pid = True
 
     # Create logs directory to ensure it exists
     log_dir = os.path.join(cfg.output_dir, "logs")
@@ -76,7 +132,7 @@ def actor_cli(cfg: TrainPipelineConfig):
     log_file = os.path.join(log_dir, f"actor_{cfg.job_name}.log")
 
     # Initialize logging with explicit log file
-    init_logging(log_file=log_file)
+    init_logging(log_file=log_file, display_pid=display_pid)
     logging.info(f"Actor logging initialized, writing to {log_file}")
 
     shutdown_event = setup_process_handlers(use_threads(cfg))
@@ -193,7 +249,7 @@ def act_with_policy(
         log_dir = os.path.join(cfg.output_dir, "logs")
         os.makedirs(log_dir, exist_ok=True)
         log_file = os.path.join(log_dir, f"actor_policy_{os.getpid()}.log")
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Actor policy process logging initialized")
 
     logging.info("make_env online")
@@ -223,12 +279,13 @@ def act_with_policy(
     # NOTE: For the moment we will solely handle the case of a single environment
     sum_reward_episode = 0
     list_transition_to_send_to_learner = []
-    list_policy_time = []
     episode_intervention = False
     # Add counters for intervention rate calculation
     episode_intervention_steps = 0
     episode_total_steps = 0
 
+    policy_timer = TimerManager("Policy inference", log=False)
+
     for interaction_step in range(cfg.policy.online_steps):
         start_time = time.perf_counter()
         if shutdown_event.is_set():
@@ -237,13 +294,9 @@ def act_with_policy(
 
         if interaction_step >= cfg.policy.online_step_before_learning:
             # Time policy inference and check if it meets FPS requirement
-            with TimerManager(
+            with policy_timer:
-                elapsed_time_list=list_policy_time,
-                label="Policy inference time",
-                log=False,
-            ) as timer:  # noqa: F841
                 action = policy.select_action(batch=obs)
-            policy_fps = 1.0 / (list_policy_time[-1] + 1e-9)
+            policy_fps = policy_timer.fps_last
 
             log_policy_frequency_issue(policy_fps=policy_fps, cfg=cfg, interaction_step=interaction_step)
 
@@ -291,8 +344,8 @@ def act_with_policy(
                 )
                 list_transition_to_send_to_learner = []
 
-            stats = get_frequency_stats(list_policy_time)
+            stats = get_frequency_stats(policy_timer)
-            list_policy_time.clear()
+            policy_timer.reset()
 
             # Calculate intervention rate
             intervention_rate = 0.0
@@ -429,7 +482,7 @@ def receive_policy(
         log_file = os.path.join(log_dir, f"actor_receive_policy_{os.getpid()}.log")
 
         # Initialize logging with explicit log file
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Actor receive policy process logging initialized")
 
         # Setup process handlers to handle shutdown signal
@@ -484,7 +537,7 @@ def send_transitions(
         log_file = os.path.join(log_dir, f"actor_transitions_{os.getpid()}.log")
 
         # Initialize logging with explicit log file
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Actor transitions process logging initialized")
 
         # Setup process handlers to handle shutdown signal
@@ -533,7 +586,7 @@ def send_interactions(
         log_file = os.path.join(log_dir, f"actor_interactions_{os.getpid()}.log")
 
         # Initialize logging with explicit log file
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Actor interactions process logging initialized")
 
         # Setup process handlers to handle shutdown signal
@@ -632,25 +685,24 @@ def push_transitions_to_transport_queue(transitions: list, transitions_queue):
     transitions_queue.put(transitions_to_bytes(transition_to_send_to_learner))
 
 
-def get_frequency_stats(list_policy_time: list[float]) -> dict[str, float]:
+def get_frequency_stats(timer: TimerManager) -> dict[str, float]:
     """Get the frequency statistics of the policy.
 
     Args:
-        list_policy_time (list[float]): The list of policy times.
+        timer (TimerManager): The timer with collected metrics.
 
     Returns:
         dict[str, float]: The frequency statistics of the policy.
     """
     stats = {}
-    list_policy_fps = [1.0 / t for t in list_policy_time]
+    if timer.count > 1:
-    if len(list_policy_fps) > 1:
+        avg_fps = timer.fps_avg
-        policy_fps = mean(list_policy_fps)
+        p90_fps = timer.fps_percentile(90)
-        quantiles_90 = quantiles(list_policy_fps, n=10)[-1]
+        logging.debug(f"[ACTOR] Average policy frame rate: {avg_fps}")
-        logging.debug(f"[ACTOR] Average policy frame rate: {policy_fps}")
+        logging.debug(f"[ACTOR] Policy frame rate 90th percentile: {p90_fps}")
-        logging.debug(f"[ACTOR] Policy frame rate 90th percentile: {quantiles_90}")
         stats = {
-            "Policy frequency [Hz]": policy_fps,
+            "Policy frequency [Hz]": avg_fps,
-            "Policy frequency 90th-p [Hz]": quantiles_90,
+            "Policy frequency 90th-p [Hz]": p90_fps,
         }
     return stats
 

lerobot/scripts/server/crop_dataset_roi.py

@@ -203,6 +203,7 @@ def convert_lerobot_dataset_to_cropper_lerobot_dataset(
         if key in new_dataset.meta.info["features"]:
             new_dataset.meta.info["features"][key]["shape"] = [3] + list(resize_size)
 
+    # TODO:  Directly modify the mp4 video + meta info features, instead of recreating a dataset
     prev_episode_index = 0
     for frame_idx in tqdm(range(len(original_dataset))):
         frame = original_dataset[frame_idx]

lerobot/scripts/server/end_effector_control_utils.py

@@ -23,10 +23,9 @@ import numpy as np
 import torch
 
 from lerobot.common.robot_devices.utils import busy_wait
+from lerobot.common.utils.utils import init_logging
 from lerobot.scripts.server.kinematics import RobotKinematics
 
-logging.basicConfig(level=logging.INFO)
-
 
 class InputController:
     """Base class for input controllers that generate motion deltas."""
@@ -726,6 +725,8 @@ if __name__ == "__main__":
     from lerobot.common.robot_devices.robots.utils import make_robot_from_config
     from lerobot.scripts.server.gym_manipulator import make_robot_env
 
+    init_logging()
+
     parser = argparse.ArgumentParser(description="Test end-effector control")
     parser.add_argument(
         "--mode",

lerobot/scripts/server/gym_manipulator.py

@@ -1588,19 +1588,20 @@ class GamepadControlWrapper(gym.Wrapper):
             input_threshold: Minimum movement delta to consider as active input.
         """
         super().__init__(env)
-        from lerobot.scripts.server.end_effector_control_utils import (
-            GamepadController,
-            GamepadControllerHID,
-        )
 
         # use HidApi for macos
         if sys.platform == "darwin":
+            # NOTE: On macOS, pygame doesn’t reliably detect input from some controllers so we fall back to hidapi
+            from lerobot.scripts.server.end_effector_control_utils import GamepadControllerHID
+
             self.controller = GamepadControllerHID(
                 x_step_size=x_step_size,
                 y_step_size=y_step_size,
                 z_step_size=z_step_size,
             )
         else:
+            from lerobot.scripts.server.end_effector_control_utils import GamepadController
+
             self.controller = GamepadController(
                 x_step_size=x_step_size,
                 y_step_size=y_step_size,
@@ -1748,6 +1749,8 @@ class GymHilDeviceWrapper(gym.Wrapper):
         for k in obs:
             obs[k] = obs[k].to(self.device)
         if "action_intervention" in info:
+            # NOTE: This is a hack to ensure the action intervention is a float32 tensor and supported on MPS device
+            info["action_intervention"] = info["action_intervention"].astype(np.float32)
             info["action_intervention"] = torch.from_numpy(info["action_intervention"]).to(self.device)
         return obs, reward, terminated, truncated, info
 
@@ -1756,6 +1759,8 @@ class GymHilDeviceWrapper(gym.Wrapper):
         for k in obs:
             obs[k] = obs[k].to(self.device)
         if "action_intervention" in info:
+            # NOTE: This is a hack to ensure the action intervention is a float32 tensor and supported on MPS device
+            info["action_intervention"] = info["action_intervention"].astype(np.float32)
             info["action_intervention"] = torch.from_numpy(info["action_intervention"]).to(self.device)
         return obs, info
 

lerobot/scripts/server/learner_server.py

@@ -14,6 +14,66 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+"""
+Learner server runner for distributed HILSerl robot policy training.
+
+This script implements the learner component of the distributed HILSerl architecture.
+It initializes the policy network, maintains replay buffers, and updates
+the policy based on transitions received from the actor server.
+
+Examples of usage:
+
+- Start a learner server for training:
+```bash
+python lerobot/scripts/server/learner_server.py --config_path lerobot/configs/train_config_hilserl_so100.json
+```
+
+- Run with specific SAC hyperparameters:
+```bash
+python lerobot/scripts/server/learner_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --learner.sac.alpha=0.1 \
+    --learner.sac.gamma=0.99
+```
+
+- Run with a specific dataset and wandb logging:
+```bash
+python lerobot/scripts/server/learner_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --dataset.repo_id=username/pick_lift_cube \
+    --wandb.enable=true \
+    --wandb.project=hilserl_training
+```
+
+- Run with a pretrained policy for fine-tuning:
+```bash
+python lerobot/scripts/server/learner_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --pretrained_policy_name_or_path=outputs/previous_training/checkpoints/080000/pretrained_model
+```
+
+- Run with a reward classifier model:
+```bash
+python lerobot/scripts/server/learner_server.py \
+    --config_path lerobot/configs/train_config_hilserl_so100.json \
+    --reward_classifier_pretrained_path=outputs/reward_model/best_model
+```
+
+**NOTE**: Start the learner server before launching the actor server. The learner opens a gRPC server
+to communicate with actors.
+
+**NOTE**: Training progress can be monitored through Weights & Biases if wandb.enable is set to true
+in your configuration.
+
+**WORKFLOW**:
+1. Create training configuration with proper policy, dataset, and environment settings
+2. Start this learner server with the configuration
+3. Start an actor server with the same configuration
+4. Monitor training progress through wandb dashboard
+
+For more details on the complete HILSerl training workflow, see:
+https://github.com/michel-aractingi/lerobot-hilserl-guide
+"""
 
 import logging
 import os
@@ -73,7 +133,6 @@ from lerobot.scripts.server.utils import (
 
 LOG_PREFIX = "[LEARNER]"
 
-logging.basicConfig(level=logging.INFO)
 
 #################################################
 # MAIN ENTRY POINTS AND CORE ALGORITHM FUNCTIONS #
@@ -113,13 +172,17 @@ def train(cfg: TrainPipelineConfig, job_name: str | None = None):
     if job_name is None:
         raise ValueError("Job name must be specified either in config or as a parameter")
 
+    display_pid = False
+    if not use_threads(cfg):
+        display_pid = True
+
     # Create logs directory to ensure it exists
     log_dir = os.path.join(cfg.output_dir, "logs")
     os.makedirs(log_dir, exist_ok=True)
     log_file = os.path.join(log_dir, f"learner_{job_name}.log")
 
     # Initialize logging with explicit log file
-    init_logging(log_file=log_file)
+    init_logging(log_file=log_file, display_pid=display_pid)
     logging.info(f"Learner logging initialized, writing to {log_file}")
     logging.info(pformat(cfg.to_dict()))
 
@@ -275,7 +338,7 @@ def add_actor_information_and_train(
         log_dir = os.path.join(cfg.output_dir, "logs")
         os.makedirs(log_dir, exist_ok=True)
         log_file = os.path.join(log_dir, f"learner_train_process_{os.getpid()}.log")
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Initialized logging for actor information and training process")
 
     logging.info("Initializing policy")
@@ -604,7 +667,7 @@ def start_learner_server(
         log_file = os.path.join(log_dir, f"learner_server_process_{os.getpid()}.log")
 
         # Initialize logging with explicit log file
-        init_logging(log_file=log_file)
+        init_logging(log_file=log_file, display_pid=True)
         logging.info("Learner server process logging initialized")
 
         # Setup process handlers to handle shutdown signal

pyproject.toml

@@ -84,7 +84,7 @@ dora = [
 ]
 dynamixel = ["dynamixel-sdk>=3.7.31", "pynput>=1.7.7"]
 feetech = ["feetech-servo-sdk>=1.0.0", "pynput>=1.7.7"]
-hilserl = ["transformers>=4.48.0", "gym-hil>=0.1.2", "protobuf>=5.29.3", "grpcio>=1.70.0"]
+hilserl = ["transformers>=4.48", "gym-hil>=0.1.3", "protobuf>=5.29.3", "grpcio>=1.70.0"]
 intelrealsense = ["pyrealsense2>=2.55.1.6486 ; sys_platform != 'darwin'"]
 pi0 = ["transformers>=4.48.0"]
 pusht = ["gym-pusht>=0.1.5 ; python_version < '4.0'"]

tests/policies/test_sac_config.py

@@ -59,7 +59,6 @@ def test_sac_config_default_initialization():
     assert config.num_critics == 2
 
     # Architecture specifics
-    assert config.camera_number == 1
     assert config.vision_encoder_name is None
     assert config.freeze_vision_encoder is True
     assert config.image_encoder_hidden_dim == 32

tests/policies/test_sac_policy.py

@@ -9,6 +9,13 @@ from lerobot.common.policies.sac.modeling_sac import MLP, SACPolicy
 from lerobot.common.utils.random_utils import seeded_context
 from lerobot.configs.types import FeatureType, PolicyFeature
 
+try:
+    import transformers  # noqa: F401
+
+    TRANSFORMERS_AVAILABLE = True
+except ImportError:
+    TRANSFORMERS_AVAILABLE = False
+
 
 def test_mlp_with_default_args():
     mlp = MLP(input_dim=10, hidden_dims=[256, 256])
@@ -274,6 +281,7 @@ def test_sac_policy_with_visual_input(batch_size: int, state_dim: int, action_di
     "batch_size,state_dim,action_dim,vision_encoder_name",
     [(1, 6, 6, "helper2424/resnet10"), (1, 6, 6, "facebook/convnext-base-224")],
 )
+@pytest.mark.skipif(not TRANSFORMERS_AVAILABLE, reason="Transformers are not installed")
 def test_sac_policy_with_pretrained_encoder(
     batch_size: int, state_dim: int, action_dim: int, vision_encoder_name: str
 ):