From a0570c6eab551ec34792d021f491a8818dcd04ef Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=95=D0=B2=D0=B3=D0=B5=D0=BD=D0=B8=D1=8F=20=D0=A1=D1=83?= =?UTF-8?q?=D1=85=D0=BE=D0=B4=D0=BE=D0=BB=D1=8C=D1=81=D0=BA=D0=B0=D1=8F?= Date: Thu, 30 Jan 2025 18:26:11 +0100 Subject: [PATCH] review --- .../documentation/fastembed/fastembed-rerankers.md | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/qdrant-landing/content/documentation/fastembed/fastembed-rerankers.md b/qdrant-landing/content/documentation/fastembed/fastembed-rerankers.md index 8e49e4d54..369c199a2 100644 --- a/qdrant-landing/content/documentation/fastembed/fastembed-rerankers.md +++ b/qdrant-landing/content/documentation/fastembed/fastembed-rerankers.md @@ -13,16 +13,18 @@ Rerankers analyze in-depth token-level interactions between the query and each d ## Goal of this Tutorial -It's common to use [cross-enconder]((https://sbert.net/examples/applications/cross-encoder/README.html)) models as rerankers. This tutorial uses [Jina Reranker v2 Base Multilingual](https://jina.ai/news/jina-reranker-v2-for-agentic-rag-ultra-fast-multilingual-function-calling-and-code-search/) -- cross-encoder reranker supported in FastEmbed. +It's common to use [cross-enconder](https://sbert.net/examples/applications/cross-encoder/README.html) models as rerankers. This tutorial uses [Jina Reranker v2 Base Multilingual](https://jina.ai/news/jina-reranker-v2-for-agentic-rag-ultra-fast-multilingual-function-calling-and-code-search/) (licensed under CC-BY-NC-4.0) -- cross-encoder reranker supported in FastEmbed. + + We use the `all-MiniLM-L6-v2` dense embedding model (also supported in FastEmbed) as a first-stage retriever and then refine results with `Jina Reranker v2`. ## Setup -Install `fastembed`. +Install `qdrant-client` with `fastembed`. ```python -pip install fastembed +pip install "qdrant-client[fastembed]" ``` Imports cross-encoders and text embeddings for the first-stage retrieval. @@ -134,12 +136,6 @@ descriptions_embeddings = list( Let's upload embeddings to Qdrant. -Install `qdrant-client` - -```python -pip install qdrant-client -``` - Qdrant Client has a simple in-memory mode that allows you to experiment locally on small data volumes. Alternatively, you could use for experiments [a free cluster](https://qdrant.tech/documentation/cloud/create-cluster/#create-a-cluster) in Qdrant Cloud.