File size: 4,200 Bytes
0d80452 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 | #!/usr/bin/env python
# Copyright 2025 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# 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.
from dataclasses import dataclass
from lerobot.configs import PipelineFeatureType, PolicyFeature
from lerobot.types import EnvAction, EnvTransition, PolicyAction, TransitionKey
from .converters import to_tensor
from .hil_processor import TELEOP_ACTION_KEY
from .pipeline import ActionProcessorStep, ProcessorStep, ProcessorStepRegistry
@ProcessorStepRegistry.register("torch2numpy_action_processor")
@dataclass
class Torch2NumpyActionProcessorStep(ActionProcessorStep):
"""
Converts a PyTorch tensor action to a NumPy array.
This step is useful when the output of a policy (typically a torch.Tensor)
needs to be passed to an environment or component that expects a NumPy array.
Attributes:
squeeze_batch_dim: If True, removes the first dimension of the array
if it is of size 1. This is useful for converting a
batched action of size (1, D) to a single action of size (D,).
"""
squeeze_batch_dim: bool = True
def action(self, action: PolicyAction) -> EnvAction:
if not isinstance(action, PolicyAction):
raise TypeError(
f"Expected PolicyAction or None, got {type(action).__name__}. "
"Use appropriate processor for non-tensor actions."
)
numpy_action = action.detach().cpu().numpy()
# Remove batch dimensions but preserve action dimensions.
# Only squeeze if there's a batch dimension (first dim == 1).
if (
self.squeeze_batch_dim
and numpy_action.shape
and len(numpy_action.shape) > 1
and numpy_action.shape[0] == 1
):
numpy_action = numpy_action.squeeze(0)
return numpy_action
def transform_features(
self, features: dict[PipelineFeatureType, dict[str, PolicyFeature]]
) -> dict[PipelineFeatureType, dict[str, PolicyFeature]]:
return features
@ProcessorStepRegistry.register("numpy2torch_action_processor")
@dataclass
class Numpy2TorchActionProcessorStep(ProcessorStep):
"""Converts a NumPy array action to a PyTorch tensor when action is present."""
def __call__(self, transition: EnvTransition) -> EnvTransition:
"""Converts numpy action to torch tensor if action exists, otherwise passes through."""
self._current_transition = transition.copy()
new_transition = self._current_transition
action = new_transition.get(TransitionKey.ACTION)
if action is not None:
if not isinstance(action, EnvAction):
raise TypeError(
f"Expected np.ndarray or None, got {type(action).__name__}. "
"Use appropriate processor for non-tensor actions."
)
torch_action = to_tensor(action, dtype=None) # Preserve original dtype
new_transition[TransitionKey.ACTION] = torch_action
complementary_data = new_transition.get(TransitionKey.COMPLEMENTARY_DATA, {})
if TELEOP_ACTION_KEY in complementary_data:
teleop_action = complementary_data[TELEOP_ACTION_KEY]
if isinstance(teleop_action, EnvAction):
complementary_data[TELEOP_ACTION_KEY] = to_tensor(teleop_action)
new_transition[TransitionKey.COMPLEMENTARY_DATA] = complementary_data
return new_transition
def transform_features(
self, features: dict[PipelineFeatureType, dict[str, PolicyFeature]]
) -> dict[PipelineFeatureType, dict[str, PolicyFeature]]:
return features
|