IndexNode in Joint Tabular/Semantic QA over Tesla 10K

[robjm16]

I am trying to run this example notebook and get stuck on this line.

example_index_node = [b for b in base_nodes_2021 if isinstance(b, IndexNode)][20]

It returns an empty list, as I think that none of the base_nodes_2021 are recognized as an IndexNode.

I am new to llama_index.

Any ideas?

[dosubot[bot]]

🤖

Hey @robjm16! 👋

I'm Dosu, a helpful bot here to assist you while we wait for a human maintainer. I'm pretty good at squashing bugs, finding answers, and helping users like you become contributors. Just think of me as your friendly digital hand-raiser. 🤖

Don't worry, we'll get your issue figured out in no time! 😊

It seems like you're having trouble with the IndexNode class and the base_nodes_2021 list. Based on the information available in the LlamaIndex repository, it's not entirely clear what the IndexNode class is or how it's related to the BaseNode class.

However, I can tell you that the base_nodes_2021 list should be a list of instances of the BaseNode class or a subclass of BaseNode. Here's an example of how you might initialize this list:

from llama_index.schema import BaseNode

# Create your nodes
base_nodes_2021 = [BaseNode(id='node1', name='Node 1'), BaseNode(id='node2', name='Node 2')]

In the line of code you're having trouble with, it's trying to find the first 20 instances in base_nodes_2021 that are instances of IndexNode. If IndexNode is a subclass of BaseNode, and base_nodes_2021 contains instances of IndexNode, this line of code should work. However, if base_nodes_2021 doesn't contain any instances of IndexNode, this line of code will indeed return an empty list.

Without more information about the IndexNode class and how your base_nodes_2021 list is being populated, it's hard to provide a more specific solution. If you could provide more details about these, it would be helpful in diagnosing the issue. For example, how are you populating base_nodes_2021? And can you confirm whether IndexNode is a subclass of BaseNode?

You can refer to the BaseIndex class in the LlamaIndex repository for more information on how BaseNode instances are used.

Sources

• llama_index/indices/base.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[robjm16]

This is the code that precedes the problematic line:

from llama_index.readers.file.flat_reader import FlatReader
from pathlib import Path

reader = FlatReader()
docs_2021 = reader.load_data(Path("tesla_2021_10k.htm"))
docs_2020 = reader.load_data(Path("tesla_2020_10k.htm"))
from llama_index.node_parser import (
UnstructuredElementNodeParser,
)

node_parser = UnstructuredElementNodeParser()
import os
import pickle

if not os.path.exists("2021_nodes.pkl"):
raw_nodes_2021 = node_parser.get_nodes_from_documents(docs_2021)
pickle.dump(raw_nodes_2021, open("2021_nodes.pkl", "wb"))
else:
raw_nodes_2021 = pickle.load(open("2021_nodes.pkl", "rb"))
100%|██████████████████████████████████████████████████████████████████| 105/105 [14:59<00:00, 8.56s/it]
base_nodes_2021, node_mappings_2021 = node_parser.get_base_nodes_and_mappings(
raw_nodes_2021
)