API Initialization Issue

[VEDANTBISHT]

After running the code from store_index.py file

pinecone.init(api_key=PINECONE_API_KEY, environment=PINECONE_API_ENV)

It shows the given error:-

Traceback (most recent call last): File "C:\Users\vedan\Downloads\kuch bhi\End-to-end-Medical-Chatbot-using-Llama2\store_index.py", line 21, in pinecone.init(api_key=PINECONE_API_KEY, File "C:\Users\vedan\AppData\Roaming\Python\Python312\site-packages\pinecone\deprecation_warnings.py", line 38, in init raise AttributeError(msg) AttributeError: init is no longer a top-level attribute of the pinecone package.

Please create an instance of the Pinecone class instead.

Example:

import os
from pinecone import Pinecone, ServerlessSpec

pc = Pinecone(
    api_key=os.environ.get("PINECONE_API_KEY")
)

# Now do stuff
if 'my_index' not in pc.list_indexes().names():
    pc.create_index(
        name='my_index',
        dimension=1536,
        metric='euclidean',
        spec=ServerlessSpec(
            cloud='aws',
            region='us-west-2'
        )
    )

Can you pls help me find the error as init is no longer supported by Pinecone.

[smriti217]

Is it working now?

[dkk3675]

@VEDANTBISHT I was facing the same issue. After debugging for much time, got to know that pinecone-client version <=2.2.4 only supports init package and is deprecated in furthur releases. [Ref. : Pinecone-Migration-Guide]

[TenzinNsut]

Uninstall the current version of pinecone

pip uninstall pinecone-client

Now install pinecone version: 2.2.4

pip install pinecone-client==2.2.4

[nortln]

I reduced to a lower version but after that there was no more environment variable on the pinecone website and when I tried using chromadb it didn't work, especially with the hugging face embeddings I have been trying to solve this but to no avail. Does anyone have a fix

[d-pamneja]

I encountered the same issue, this can be tackled as follows.

Since __init__ is no longer a top-level attribute of the pinecone package, we will have to create an instance of pinecone via the pinecone library itself as follows.

from pinecone import Pinecone

# Creating a Pinecone instance
pc = Pinecone(api_key = os.environ["PINECONE_API_KEY"], environment = os.environ["PINECONE_API_ENV"])

index_name = "XYZ"

Now, when you would try to add the text chunks to the given index, it would again throw an error saying list_indexes() are no longer a top-level attribute of the pinecone package, as the from_texts() method would try to use this functionality and throw an error. To work around that, we will have to import Pinecone again, but this time using the langchain_pinecone library. The same can be seen as shown below: