google/generativeai/embedding.py [171:197]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    if title and to_task_type(task_type) is not EmbeddingTaskType.RETRIEVAL_DOCUMENT:
        raise ValueError(
            f"Invalid task type: When a title is specified, the task must be of a 'retrieval document' type. Received task type: {task_type} and title: {title}."
        )

    if output_dimensionality and output_dimensionality < 0:
        raise ValueError(
            f"Invalid value: `output_dimensionality` must be a non-negative integer. Received: {output_dimensionality}."
        )

    if task_type:
        task_type = to_task_type(task_type)

    if isinstance(content, Iterable) and not isinstance(content, (str, Mapping)):
        result = {"embedding": []}
        requests = (
            protos.EmbedContentRequest(
                model=model,
                content=content_types.to_content(c),
                task_type=task_type,
                title=title,
                output_dimensionality=output_dimensionality,
            )
            for c in content
        )
        for batch in _batched(requests, EMBEDDING_MAX_BATCH_SIZE):
            embedding_request = protos.BatchEmbedContentsRequest(model=model, requests=batch)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



google/generativeai/embedding.py [265:290]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    if title and to_task_type(task_type) is not EmbeddingTaskType.RETRIEVAL_DOCUMENT:
        raise ValueError(
            f"Invalid task type: When a title is specified, the task must be of a 'retrieval document' type. Received task type: {task_type} and title: {title}."
        )
    if output_dimensionality and output_dimensionality < 0:
        raise ValueError(
            f"Invalid value: `output_dimensionality` must be a non-negative integer. Received: {output_dimensionality}."
        )

    if task_type:
        task_type = to_task_type(task_type)

    if isinstance(content, Iterable) and not isinstance(content, (str, Mapping)):
        result = {"embedding": []}
        requests = (
            protos.EmbedContentRequest(
                model=model,
                content=content_types.to_content(c),
                task_type=task_type,
                title=title,
                output_dimensionality=output_dimensionality,
            )
            for c in content
        )
        for batch in _batched(requests, EMBEDDING_MAX_BATCH_SIZE):
            embedding_request = protos.BatchEmbedContentsRequest(model=model, requests=batch)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -