Explicitly controls the probability of dropping out subsets of A, B, C (action, text, image). Each item in the batch is mapped to exactly one

A fast approximation of the cumulative distribution function of the standard normal.

Given a dictionary of trajectories where each key is a trajectory identifier (e.g., a folder path) and each value is a dictionary mapping "St

Defines how much the task base can rotate during episode placement. Default is set such that it can rotate any amount on the z axis.

Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of (1-beta) over time from t = [

Compute the distance matrix between two lists of 3D points. points_a: List of np.array([x, y, z]) with shape (3,). Length N.

checks if a loaded state_dict component is part of quantized param + some validation; only defined for quantization methods that requ

Cleans up all detected objects and resets the object ID.

Main data collection loop.

Based on the input mask corresponding value `0 or 1` for each phrases and image, mask the features corresponding to phrases and image

Join scene description with candidate plans; preserves original spacing between sentences.

Compute what the p_mean_variance output would have been, should the model's score function be conditioned by cond_fn. See con

Convert a list of action IDs to their corresponding strings.

Takes a state dict and a config, and returns a converted checkpoint.

Creates a dynamic module in the cache directory for modules.

Creates a config for the diffusers based on the config of the LDM model.

r""" Denormalize an image array to [0,1]. Args: images (`np.ndarray` or `torch.Tensor`): The image array

主函数：遍历每张图，先投影，再绘制，再保存。

Draws a bounding box around the largest connected component of the target in sem_obs, then scales the box by the provided scale factor. If no

Enable adapters that are attached to the model. The model uses `self.active_adapters()` to retrieve the list of adapters to enable.