Creates a Torch function that applies the given function to each entry of a TextTensor.

Retrieves embeddings for the given texts from the OpenAI API and saves the results in a file. Parameters: texts (List[str], TextTens

Joins all 'Str' types to each other if the second item in the tuple is a string. Args: pandoc_tuples (list): The list of Pandoc tupl

Count the number of letters in the request. Note that this is a naive implementation and might not reflect the actual cost of a request on GCP's V

Count the number of tokens in the request. Only supports completion and embedding requests.

Converts a given string into a Pandoc AST JSON format. Args: input_string (str): The input string to be converted. input_for

Maps a description of args to a dictionary of {argname: description}. Input: (' weight (Tensor): a weight tensor\n' + '

Processes API requests in parallel, throttling to stay under rate limits.

Recursively traverse the XML elements and build a tuple structure.

Performs a single optimization step. Args: closure (callable, optional): A closure that reevaluates the model and ret

Formats a TextTensor into a matrix representation with entries padded with spaces to be able to have multi-line entries aligned. It would be p

Converts an XML string to a structured tuple representation of its AST.

Get item(s) from the TextTensor based on the key. Supports nested access using dot notation. :param k

TextualGradientDescent optimizer :param engine: the engine to use for updating variables :type engine: EngineLM :param parame