Python OpenAI API: Can't instantiate abstract class CustomExtractor with abstract method class_name

Question

I am following this example from the llama_index docs: https://github.com/run-llama/llama_index/blob/main/docs/examples/metadata_extraction/MetadataExtractionSEC.ipynb to use a custom metadata extractor with my code like so:

from llama_index.node_parser import SimpleNodeParser
from llama_index.node_parser.extractors import (
    MetadataExtractor,
    MetadataFeatureExtractor,
)

class CustomExtractor(MetadataFeatureExtractor):
    def extract(self, nodes):
        metadata_list = [
            {
                "custom": node.metadata["document_title"]
                + "\n"
                + node.metadata["excerpt_keywords"]
            }
            for node in nodes
        ]
        return metadata_list


metadata_extractor = MetadataExtractor(
    extractors=[
        CustomExtractor()
    ],
)

But running the code fails with the following error:

Can't instantiate abstract class CustomExtractor with abstract method class_name

I don't believe there is a syntax error here. Any help is appreciated.

Answer 1

MetadataFeatureExtractor extends BaseExtractor which extends BaseComponent which defines an @abstractmethod called class_name(). You need to implement that in your custom extractor. Try

class CustomExtractor(MetadataFeatureExtractor):
    @classmethod
    def class_name(cls):
        return 'CustomExtractor'

    def extract(self, nodes):
        metadata_list = [
            {
                "custom": node.metadata["document_title"]
                + "\n"
                + node.metadata["excerpt_keywords"]
            }
            for node in nodes
        ]
        return metadata_list