Computes an OCR-based reward score using a remote TextPecker model served with vLLM.

Create sinusoidal timestep embeddings. :param t: a 1-D Tensor of N indices, one per batch element. These ma

Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distribution :

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

:param images: List of images or videos (each video as np.ndarray of shape [F, H, W, C]) :param prompts: List of prompts containing t

r""" Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[str]`, *optional*):

r""" Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[str]`, *optional*):

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

Calculate OCR reward :param images: List of input images (PIL or numpy format) :param prompts: Corresponding target text list

:param images: List of images or videos (each video as np.ndarray of shape [F, H, W, C]) :param prompts: List of prompts containing t

Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments