IVYXSTUDIO · IVY NODE

Question

ivy.node.question · v0.0.0

ivyx✓

Converts a textual question into a numerical vector representation using a text vectorizer. This node is essential for RAG (Retrieval-Augmented Generation) pipelines where questions need to be vectorized for similarity search. The output vectorize array is used with milvus-search to find similar content, while the text output preserves the original question. Requires text-vectorizer from text-vectorizer node. Typically used before milvus-search in question-answering workflows.

#Question

Inputs

Field	Type	Description
text_vectorizerrequired	object	A text vectorizer instance responsible for generating embeddings (Node reference)
questionrequired	string	The question to be vectorized
features	object	Features
labels	object	Labels

Outputs

Field	Type	Description
vectorizerequired	array	Not declared
textrequired	string	Not declared
features	object	Features
labels	object	Labels

Source

python

# Input preparation
inp = __ivy_ctx__["nodes"][__ivy_node_id__]["input"]
text_vectorizer = inp["text_vectorizer"]
question = inp["question"]

# Compute
from typing import List

class Question:
    """
    A class for converting a textual question into a numerical vector representation 
    using a specified text vectorizer.

    Attributes:
        text_vectorizer: An instance of a text vectorization class that provides a `vectorizer_text` method.

    Methods:
        vectorize(question: str) -> List[float]:
            Converts the given question into a numerical vector representation.
    """

    def __init__(self, text_vectorizer: TextVectorizer):
        """
        Initializes the Question class with a text vectorizer.

        Args:
            text_vectorizer (TextVectorizer): An instance of a text vectorization class that 
                                              must have a callable `vectorizer_text` method.

        Raises:
            TypeError: If `text_vectorizer` does not have a callable `vectorizer_text` method.
        """
        if not hasattr(text_vectorizer, "vectorizer_text") or not callable(text_vectorizer.vectorizer_text):
            raise TypeError("text_vectorizer must be an instance of a class with a callable 'vectorizer_text' method.")
        
        self.text_vectorizer = text_vectorizer

    def vectorize(self, question: str) -> List[float]:
        """
        Converts the given question into a vector representation.

        Args:
            question (str): The input question to be vectorized.

        Returns:
            List[float]: A numerical vector representing the given question.

        Raises:
            ValueError: If `question` is not a valid non-empty string.
            RuntimeError: If vectorization fails due to an internal error.
        """
        if not isinstance(question, str):
            raise ValueError("The question must be a string.")
        if not question.strip():
            raise ValueError("The question cannot be empty.")

        try:
            return self.text_vectorizer.vectorizer_text(question)
        except Exception as e:
            raise RuntimeError(f"Vectorization failed: {str(e)}")
        
question_obj = Question(text_vectorizer=text_vectorizer)

vectorize = question_obj.vectorize(question=question)

text = question

# Output collection (runner reads __ivy_ctx__)
out = __ivy_ctx__["nodes"][__ivy_node_id__]["output"]
out["vectorize"] = vectorize
out["text"] = text

Tests

Requires: python:3.11

basic-question-vectorization
Vectorize valid question (should succeed).
questionvectorizationbasic
error-empty-question
Empty question should raise an error.
error-handlingvalidation