bbornsztein

Log inLog into community

Find answers from the community

Home

Members

bbornsztein

Offline, last seen 2 months ago

Joined September 25, 2024

bbbornsztein

GPTPineconeIndex error

Getting this error when using the GPTPineconeIndex (on a rather large set of documents):

Plain Text

ValueError: Effective chunk size is non positive after considering extra_info

or others: any idea what's going on?

6 comments

bbbornsztein

Not sure how to resolve this error `The

Not sure how to resolve this error: The batch size should not be larger than 2048

6 comments

bbbornsztein

👋 I added Zendesk API Intercom API and

👋 I added Zendesk API, Intercom API, and Wordpress API document loaders to AgentHQ this morning. Happy to contribute back llama hub if there's interest.

12 comments

bbbornsztein

GitHub - abhijithneilabraham/tableQA: AI...

I was just looking at: https://github.com/abhijithneilabraham/tableQA

Not sure about the timeseries stuff, and not connected/integrated with LlamaIndex, but worth a look for QA over tabular data.

7 comments

bbbornsztein

Getting this error when running a

Getting this error when running a GPTSimpleVectorIndex:

A single term is larger than the allowed chunk size. Term size: 511 Chunk size: 512Effective chunk size: 476

What's going on there?

8 comments

bbbornsztein

Right so I m doing this and finding the

Right - so I'm doing this and finding the ListIndex is much faster with required_keyword (makes sense).

Plain Text

keywords = extract_keywords_given_response(
                input, start_token="")

query_result = index.query(
                input, verbose=True, llm_predictor=llm_predictor, required_keywords=keywords)

For large indexes I was finding it too slow to be usable. Wondering if extracting keywords should be a default (or at least suggested)

7 comments

bbbornsztein

Wondering if there s any way to save

Wondering if there's any way to save/load an index to S3 or some other filestore? I suppose we'd then need to download the whole index every time a query is run?

11 comments