GraphQL rebuild error after rolling one or more replicas
Description i’m seeing some weird behavior with the latest weaviate version running on kubernetes, where cycling pods results in one of the replicas (i have 3) reporting failures (one of them has...
View ArticleHow to filter by date field
Hi @DudaNogueira, I’m currently inserting data using LlamaIndex and I assumed I could use node.metadata to filter by dates. for obj in response.objects: print(obj.properties.keys()) #...
View ArticleClik here to view.
Why distance score not equal to 0 (searching exactly the same words)
Thank you very much for the answer. But when I follow you, why are there no output?
View ArticlePossible bug with relativeScoreFusion
IMHO the documents retrieved only via vector search must be assigned a value of 0 for the bm25 part before applying the fusion, by doing that the last bm25 result will rank up as the normalized score...
View ArticleWhy distance score not equal to 0 (searching exactly the same words)
IIRC weaviate also adds the collection name into the vectorized text. Try it again with Configure.Vectorizer.text2vec_openai( ...., vectorize_collection_name=False )
View ArticleHow to catch errors with dynamic batch import
Description I’m trying to batch import data into a collection like so: with filings_collection.batch.dynamic() as batch: for fc in tqdm(filngs_chunked): batch.add_object( properties=fc, ) I’m having...
View ArticleType object '_Generative' has no attribute 'ollama'
Hi, I am encountering an error: type object '_Generative' has no attribute 'ollama'. According to the documentation, this feature is available in version 1.25.0, and I am using version 1.25.4. I...
View ArticleClik here to view.
How to catch errors with dynamic batch import
Hello @md1639, Welcome to our community! it’s great to have you here You can consider capture more information about errors and failed objects by implementing error handling and logging during the...
View ArticleUpdate golden ragtriever?
Hi everyone, Totally nubie question. I would like to update my golden ragtriever as I am currently using v 0.3.1 but I am getting itchy losing my files. I have embedded some thousands of books using...
View ArticleDeploy verba
What is the best way to deploy verba ? Is replit good ? Or we have other options ?
View ArticleCustomize verba
Hi, I am customizing verba. But I do not see any of my git files change. Are the changes only in my python virtual environment ? If so, how can I get these changes to my replit deployment ?
View ArticlePossible bug with relativeScoreFusion
I know the algorithm is expected to have only one worst case, but before the final normalization we could assume a value of 0 for all the documents missing a BM25 score. Don’t know if it’s a...
View ArticleClik here to view.
Semantic search is not as good as I expected
@Ghattas_Salloum I am using GPT34o for metadata extraction. I used openAI embedder (text-embedding-3-small - 1536 dimensions), with good results. I am thinking of using huggingface sentence...
View ArticleClik here to view.
Possible bug with relativeScoreFusion
arielmoraes: value of 0 for all the documents missing a BM25 score Bm25 scores can be negative, this won’t work
View ArticleGetting 401 error with data insertion
Description We are using an on prem instance of weaviate. We were able to do some data insertion and retrieval but started getting 401 error and unable to identify the root cause. using multi tenancy...
View ArticleClik here to view.
Customize verba
hi! You should follow Verba’s build from source docs: GitHub GitHub - weaviate/Verba: Retrieval Augmented Generation (RAG) chatbot powered... Retrieval Augmented Generation (RAG) chatbot powered by...
View ArticleClik here to view.
Deploy verba
Hi! I believe the best way to deploy Verba is using docker: GitHub GitHub - weaviate/Verba: Retrieval Augmented Generation (RAG) chatbot powered... Retrieval Augmented Generation (RAG) chatbot powered...
View ArticleGetting 401 error with data insertion
hi @aparna_raj !! Welcome to our community! Please, when asking for support in forums, always fill in the requested infos like version, deployment, etc. Please, can you share the code used for...
View Article