Pinecone: change dummy vector

[anakin87]

Related Issues

• fixes Pinecone: dummy vector is not compatible with the new API #6931
• similar to Pinecone: Dense vectors must contain at least one non-zero value. haystack-core-integrations#300

Proposed Changes:

• I changed the value of the dummy vector
• I also refactored the code to create this dummy vector once at init time (to avoid duplication)

How did you test it?

CI

Unfortunately, our CI tests are based on an outdated mock, which is also the reason why this problem never emerged until a user reported it. (in haystack-core-integrations, things are way better)

So I reproduced the issue and tested the change locally using this code:

from haystack.utils import fetch_archive_from_http
from haystack.utils import convert_files_to_docs
from haystack.nodes import PreProcessor
from haystack.document_stores.pinecone import PineconeDocumentStore
from haystack.nodes import EmbeddingRetriever



doc_dir = "data/tutorial8"
s3_url = "https://s3.eu-central-1.amazonaws.com/deepset.ai-farm-qa/datasets/documents/preprocessing_tutorial8.zip"
fetch_archive_from_http(url=s3_url, output_dir=doc_dir)

all_docs = convert_files_to_docs(dir_path=doc_dir)

preprocessor = PreProcessor(
clean_empty_lines=True,
clean_whitespace=True,
split_by="word",
split_length=100,
split_respect_sentence_boundary=True
)

docs_default = preprocessor.process(all_docs) #create a dictionary with the data in the 'content' key

document_store = PineconeDocumentStore(api_key="FAKE-API-KEY",
        environment = "gcp-starter", index="default")


print(document_store.get_document_count())

document_store.write_documents(docs_default)

retriever = EmbeddingRetriever(
    document_store=document_store, embedding_model="sentence-transformers/multi-qa-mpnet-base-dot-v1"
)

print(retriever.retrieve("What is masking in language models?"))

Notes for the reviewer

Checklist

• I have read the contributors guidelines and the code of conduct
• I have updated the related issue with new insights and changes
• I added unit tests and updated the docstrings
• I've used one of the conventional commit types for my PR title: fix:, feat:, build:, chore:, ci:, docs:, style:, refactor:, perf:, test:.
• I documented my code
• I ran pre-commit hooks and fixed any issue

[vblagoje]

@anakin87 This is the only diff (aside from -10.0) right? Everything else is the same, no?

[anakin87]

yes!

[vblagoje]

🚢