You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I tried to run the code in elasticsearch-document-store.md and ran into some errors. I attempted to fix them but got slightly stuck. If someone could point me in the right direction, happy to open a PR.
Failed to write documents to Elasticsearch. Errors:
[{'create': {'_index': 'default', '_id': '6383dc3ed51fc90c2e45704853a7ad9b14168f4f262ac7a0b65e02c465d0bb1c', 'status': 400, 'error': {'type': 'document_parsing_exception', 'reason': "[1:15833] failed to parse: The [dense_vector] field [embedding] in doc [document with id '6383dc3ed51fc90c2e45704853a7ad9b14168f4f262ac7a0b65e02c465d0bb1c'] has a different number of dimensions [768] than defined in the mapping [1024]", 'caused_by': {'type': 'illegal_argument_exception', 'reason': "The [dense_vector] field [embedding] in doc [document with id '6383dc3ed51fc90c2e45704853a7ad9b14168f4f262ac7a0b65e02c465d0bb1c'] has a different number of dimensions [768] than defined in the mapping [1024]"}}}}]'
Other than the SentenceTransformerTextEmbedder, which of these components requires us to specify a model_name_or_path? It wasn't easy to figure out from looking at the documentation or reading the Haystack source code. 🤔
The second block of code, I'm running into the same error about a mismatch in vector index lengths. There were also a few errors with param names and such that were easy to clean up:
from elasticsearch_haystack.document_store import ElasticsearchDocumentStore
from haystack.pipeline import Pipeline
from haystack.components.embedders import SentenceTransformersTextEmbedder
from elasticsearch_haystack.embedding_retriever import ElasticsearchEmbeddingRetriever
model_name_or_path = "sentence-transformers/multi-qa-mpnet-base-dot-v1"
document_store = ElasticsearchDocumentStore(hosts = "http:https://localhost:9200")
retriever = ElasticsearchEmbeddingRetriever(document_store=document_store)
text_embedder = SentenceTransformersTextEmbedder(model_name_or_path=model_name_or_path)
query_pipeline = Pipeline()
query_pipeline.add_component("text_embedder", text_embedder)
query_pipeline.add_component("retriever", retriever)
query_pipeline.connect("text_embedder.embedding", "retriever.query_embedding")
query_pipeline.run({"text_embedder": {"text": "historical places in Instanbul"}})
The text was updated successfully, but these errors were encountered:
annthurium
changed the title
code examples in elasticsearch-document-store.md throws errors
code examples in elasticsearch-document-store.md throw errors
Dec 20, 2023
I tried to run the code in
elasticsearch-document-store.md
and ran into some errors. I attempted to fix them but got slightly stuck. If someone could point me in the right direction, happy to open a PR.The topmost block of code:
Produces an error:
Other than the
SentenceTransformerTextEmbedder
, which of these components requires us to specify amodel_name_or_path
? It wasn't easy to figure out from looking at the documentation or reading the Haystack source code. 🤔The second block of code, I'm running into the same error about a mismatch in vector index lengths. There were also a few errors with param names and such that were easy to clean up:
The text was updated successfully, but these errors were encountered: