Unable to create weaviate client

Short Course Q&A Large Language Models with Semantic Search
1

On the second lecture ( Keyword search ) by running this part of code
client = weaviate.Client(
url=os.environ[‘WEAVIATE_API_URL’],
auth_client_secret=auth_config,
additional_headers={
“X-Cohere-Api-Key”: os.environ[‘COHERE_API_KEY’],
}
)

I get a error:
WeaviateStartUpError: Weaviate did not start up in 5 seconds. Either the Weaviate URL jupyter-api-proxy.internal.dlai/rev-proxy/weaviate is wrong or Weaviate did not start up in the interval given in ‘startup_period’.

Have Also tried to replace the URL value with the value from the video ( cohere-demo-weaviate network) and get the same error.

Did the WEAVIATE_API_URL been changed or some of the others env variables?

Best regards

1 Like
2

Are you trying out the code on the deeplearning.ai platform?

3

Yes, I’m running the code directly on the deeplearning.ai platform inside the Jupyter environment provided by them.

4

Oh. I think a staff needs to support on this. The course is quite old. At least 2 years. So, something may have changed.

1 Like
5

Thanks,

Will someone response to this topic or should I write somewhere else?

6

I also meet this trouble. Did you handle this problem?

7

Hello, does someone know the resolution to this question? I have the same issue

8

I think the most plausible scenario is that the weaviate cluster has expired. It’s been over two years. This page provides an easy-to-follow quick-start on setting up a weaviate cluster. It’s quite easy to set up, just that you may not be able to replicate the search in the video as that entails setting up the exact same database.
For now, I suggest you just follow the course video and learn the concept. Afterwards, set up your own cluster and practice, pending when the course will be updated by the staff. Know that a free weaviate sandbox cluster expires in 14 days. It’s basically for learning.