Get the number of joints of the robot. Returns: int: The number of joints of the robot.

Number of episodes.

Number of episodes.

Number of samples/frames.

Number of samples/frames.

Use this function to locate and debug memory leak.

Randomly shifts images horizontally and vertically. Adapted from https://github.com/facebookresearch/drqv2

Return a mapping from dataset repo_id to a dataset index automatically created by this class. This index is incorporated as a data key in the

Return the inverse mapping if repo_id_to_index.

This function ensures that a neural network trained on data collected on a given robot can work on another robot. For instance before calibration,

将输入的张量图像保存到指定路径下。 Args: img_tensor (Tensor): 输入的张量图像，形状为 (H, W, C)，数据类型为 float32 或 uint8。 key (str): 图像保存路径的标识字符串。

Set the seed when entering a context, and restore the prior random state at exit. Example usage: ``` a = random.random() # produces som

Select a single action given environment observations. This method handles caching a history of observations and an action trajectory generat

Select a single action given environment observations. This method wraps `select_actions` in order to return one action at a time for executi

Select a single action given environment observations.

Select a single action given environment observations. This method wraps `select_actions` in order to return one action at a time for executi

The provided action is expected to be a vector.

The provided action is expected to be a vector.

Tolerance in seconds used to discard loaded frames when their timestamps are not close enough from the requested frames. It is only used when

Tolerance in seconds used to discard loaded frames when their timestamps are not close enough from the requested frames. It is only used when

Returns True if this dataset loads video frames from mp4 files. Returns False if it only loads images from png files.

Returns True if this dataset loads video frames from mp4 files. Returns False if it only loads images from png files. NOTE: Fow now,

Keys to access video frames that requires to be decoded into images. Note: It is empty if the dataset contains images only, or equal

Keys to access video frames that requires to be decoded into images. Note: It is empty if the dataset contains images only, or equal