Processes an 'info' dictionary. Subclasses must implement this method. Args: info: The input 'info' dictionary from the transitio

Get the next action from the queue. Returns: Tensor | None: The next action (action_dim,) or None if queue is empty.

Create an `EnvTransition` dictionary with sensible defaults. Args: observation: Observation dictionary. action: Action dicti

Override to method to ensure all components move together.

Return the number of cached decoders.

Saves a transition, ensuring tensors are stored on the designated storage device.

Check if the queue is empty. Returns: bool: True if no actions remain, False otherwise.

Download the object from the Huggingface Hub and instantiate it. Args: pretrained_name_or_path (`str`, `Path`):

Override to return trainable parameters from both models.

Set evaluation mode for both models.

Retrieves a processor step class from the registry by its name. Args: name: The name of the step to retrieve. Returns:

Write a value to a single motor.

Return the maximum latency or None if empty.

Fixture to provide the XVLA preprocessor for tests.

Convert an `EnvTransition` back to the canonical batch format used in LeRobot. This is the inverse of `batch_to_transition`. Args:

This function only adds the frame to the episode_buffer. Apart from images — which are written in a temporary directory — nothing is

Merge new actions into the queue. This method operates differently based on RTC mode: - RTC enabled: Replaces the queue, accounting f

Log the elapsed time.

This will save to disk the current episode in self.episode_buffer. Video encoding is handled automatically based on batch_encoding_s

Clear the cache and close file handles.

Load from a YAML file.

Create a LeRobot Dataset from scratch in order to record data.

Check if a local or remote file exists. :param path: Path/URL to a file.

Given a starting state and a step function, apply beam search to find the most likely target sequences. Returns a tuple of `

Get the world size of the given process group. .. warning:: This will always return 1 if a distributed group has not been initialize

Saves the pipeline's configuration and state to a directory. This method creates a JSON configuration file that defines the pipeline's struct

Get the number of remaining actions in the queue. Returns: int: Number of unconsumed actions.

Update the running statistics with a batch of vectors. Args: batch: An array where all dimensions except the last are batch dimen

Convert various data types to PyTorch tensors with configurable options. This is a unified tensor conversion function using single dispatch

Save to a YAML file.

Track debug information for a denoising step at a given time. If a step with the given time already exists, it will be updated with the new d

Close the parquet writers. This function needs to be called after data collection/conversion, else footer metadata won't be written to the pa

Normalize a path/URL. :param path: The path/URL to normalize.

Builds a HTML table visualization of the predictions

PolicyFeature factory

Save object in local directory. Args: save_directory (`str` or `Path`): Path to directory in which the o

Establish communication with the robot. Args: calibrate (bool): If True, automatically calibrate the robot after connect

This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this

Check if a path is a URL. :param path: Path-like object to check.

Could this question be counting question that the model should use pointing for? This check is conservative, so it will have a high recall

Set seed for reproducibility.

Close the encoder, canceling any in-progress episode.

Initialize logging configuration for LeRobot. In multi-GPU training, only the main process logs to console to avoid duplicate output. Non-mai

Create or load pre- and post-processor pipelines for a given policy. This function acts as a factory. It can either load existing processor

Disconnect from the robot and perform any necessary cleanup.

Upload model checkpoint to the Hub. Use `allow_patterns` and `ignore_patterns` to precisely filter which files should be pushed to t

Write something to a file in a local or remote directory. :param dir: The path/URL of the directory to write the file to. :param fname:

Sample a random batch of transitions and collate them into batched tensors.

Set training mode for both models.

Normalize stage+tau reward to [0, 1] with custom breakpoints. Maps stage index + within-stage tau to normalized progress [0, 1]. The bre

Returns whether the question could be a counting question that would use pointing for This check is conservative by design, so it will have