From 6dbe7e7b04dbacdcc13c7c66b6c0fdca679f3172 Mon Sep 17 00:00:00 2001 From: toyeshhm Date: Tue, 25 Aug 2026 14:37:21 -0500 Subject: [PATCH 1/2] docs: fix typos across articles and building blocks Eight spelling corrections, one word each: - `Euclidian` -> `Euclidean` in the movie recommendation article - `funtionality`, `everytime`, `defualt`, `publically`, `approriate` - `Initalizing` and `classs` in the scaling-RAG code comments Line endings left as they are; the touched files use CRLF. --- docs/articles/movie_recommendation_using_vectordb.md | 2 +- docs/articles/multi-attribute-semantic-search.md | 2 +- .../retrieval_augmented_generation_eval_qdrant_ragas.md | 2 +- docs/articles/scaling_rag_for_production.md | 4 ++-- docs/articles/superlinked_langchain_retriever.md | 2 +- docs/building_blocks/vector_compute/embedding_models.md | 2 +- docs/tools/vdb_table/README.md | 2 +- 7 files changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/articles/movie_recommendation_using_vectordb.md b/docs/articles/movie_recommendation_using_vectordb.md index ab2a5dcdd..fa870d8ce 100644 --- a/docs/articles/movie_recommendation_using_vectordb.md +++ b/docs/articles/movie_recommendation_using_vectordb.md @@ -267,7 +267,7 @@ torch.save(model.state_dict(), 'trained_model.pth') ## Implementing our movie RecSys -To complete our movie recommendation system, we next set up our system so that users can input a movie title as a query. After this, our system retrieves Doc2Vec embeddings from our vector database, then uses similarity metrics such as cosine similarity or determines smallest Euclidian distances - to identify and recommend 'n' number of movies whose embeddings closely resemble those of the user's query movie, based on its genre/s. +To complete our movie recommendation system, we next set up our system so that users can input a movie title as a query. After this, our system retrieves Doc2Vec embeddings from our vector database, then uses similarity metrics such as cosine similarity or determines smallest Euclidean distances - to identify and recommend 'n' number of movies whose embeddings closely resemble those of the user's query movie, based on its genre/s. ### Setting up our vector DB diff --git a/docs/articles/multi-attribute-semantic-search.md b/docs/articles/multi-attribute-semantic-search.md index 28e9e534d..22ef9fa81 100644 --- a/docs/articles/multi-attribute-semantic-search.md +++ b/docs/articles/multi-attribute-semantic-search.md @@ -7,7 +7,7 @@ There are two common approaches to multi-attribute vector search. Both start by separately embedding each attribute of a data object. The main difference between these two approaches is in how our embeddings are *stored* and *searched*. 1. the *naive* approach - store each attribute vector in separate vector stores (one per attribute), perform a separate search for each attribute, combine search results, and post-process (e.g., weight) as required. -2. the *Superlinked* approach - concatenate and store all attribute vectors in the same vector store (using Superlinked's built-in funtionality), which allows us to *search just once*, with attendant efficiency gains. Superlinked's `spaces` *also* let us weight each attribute at query time to surface more relevant results, with no post-processing. +2. the *Superlinked* approach - concatenate and store all attribute vectors in the same vector store (using Superlinked's built-in functionality), which allows us to *search just once*, with attendant efficiency gains. Superlinked's `spaces` *also* let us weight each attribute at query time to surface more relevant results, with no post-processing. ![Two approaches to multi-attribute vector search](../assets/use_cases/multi-attribute-semantic-search/graphics.png) diff --git a/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md b/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md index 102f3f313..1d4761633 100644 --- a/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md +++ b/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md @@ -251,7 +251,7 @@ EVAL_SIZE = 10 RETRIEVAL_SIZE_3 = 3 ## The dataset used to evaluate RAG using RAGAS -## Note this is the dataset needed for evaluation hence has to be recreated everytime changes to RAG config is made +## Note this is the dataset needed for evaluation hence has to be recreated every time changes to RAG config is made rag_eval_dataset_512_3 = create_eval_dataset(qdrant_qna_dataset,EVAL_SIZE,RETRIEVAL_SIZE_3) # The dataset is then exported as a CSV file, with a filename that includes details of the experiment for easy identification, such as the chunk size along with retrieval window used in this case rag_response_dataset_512_3 = Dataset.from_dict(rag_eval_dataset_512_3) diff --git a/docs/articles/scaling_rag_for_production.md b/docs/articles/scaling_rag_for_production.md index 0df0171e5..406f6da1d 100644 --- a/docs/articles/scaling_rag_for_production.md +++ b/docs/articles/scaling_rag_for_production.md @@ -290,7 +290,7 @@ Now that our chunks are embedded, we need to **store** them somewhere. For the s from qdrant_client import QdrantClient from qdrant_client.http.models import Distance, VectorParams -# Initalizing a local client in-memory +# Initializing a local client in-memory client = QdrantClient(":memory:") client.recreate_collection( @@ -313,7 +313,7 @@ from qdrant_client.models import PointStruct def store_results(df, collection_name="documents", client=client): # Defining our data structure points = [ - # PointStruct is the data classs used in Qdrant + # PointStruct is the data class used in Qdrant PointStruct( id=hash(path), # Unique ID for each point vector=embedding, diff --git a/docs/articles/superlinked_langchain_retriever.md b/docs/articles/superlinked_langchain_retriever.md index 3dd0d0f8b..d9486ea10 100644 --- a/docs/articles/superlinked_langchain_retriever.md +++ b/docs/articles/superlinked_langchain_retriever.md @@ -188,7 +188,7 @@ These filters restrict the search space to specific filing types (e.g., 10-K, 8- .limit(sl.Param("limit")) ``` This clause tells Superlinked which fields to return in the results. This minimizes unnecessary payload and keeps downstream processing efficient by including only the necessary metadata and report content. The limit clause limits the number of retrieved documents. -Here is the complete query with defualt values for query parameters: +Here is the complete query with default values for query parameters: ```python superlinked_query = ( diff --git a/docs/building_blocks/vector_compute/embedding_models.md b/docs/building_blocks/vector_compute/embedding_models.md index 3e3784181..9e92e64d8 100644 --- a/docs/building_blocks/vector_compute/embedding_models.md +++ b/docs/building_blocks/vector_compute/embedding_models.md @@ -40,7 +40,7 @@ However, pre-trained models have performed better than custom models on others Let’s look at some examples: -[Llama-2](https://ai.meta.com/llama/), developed by Meta, is a set of LLMs pre-trained on a publically available corpus of data, with variants trained on 7B, 13B, 34B and 70B parameters. Llama-2 achieves highly impressive results on language tasks, but [as a decoder model](https://magazine.sebastianraschka.com/p/understanding-encoder-and-decoder) it is not well suited to vector embeddings. +[Llama-2](https://ai.meta.com/llama/), developed by Meta, is a set of LLMs pre-trained on a publicly available corpus of data, with variants trained on 7B, 13B, 34B and 70B parameters. Llama-2 achieves highly impressive results on language tasks, but [as a decoder model](https://magazine.sebastianraschka.com/p/understanding-encoder-and-decoder) it is not well suited to vector embeddings. Another example is OpenAI, which leverages ELMo and GPT for unsupervised pre-training to create robust general linguistic representations. Read how [OpenAI has improved language understanding with unsupervised learning](https://openai.com/blog/language-unsupervised/). diff --git a/docs/tools/vdb_table/README.md b/docs/tools/vdb_table/README.md index e3dec9ab5..910813fc0 100644 --- a/docs/tools/vdb_table/README.md +++ b/docs/tools/vdb_table/README.md @@ -37,7 +37,7 @@ Thanks for your interest in contributing to [vdbs.superlinked.com](https://vdbs. We use [discussions](https://github.com/superlinked/VectorHub/discussions/categories/vdb-comparison) as our way to have conversations about each vendor. Please find the relevant discussion and add to the conversation. -Kindly review the following sections before you submit your issue or initial pull request, and use the approriate issues/PR template. In addition, check for existing open issues and pull requests to ensure that someone else has not already corrected the information. +Kindly review the following sections before you submit your issue or initial pull request, and use the appropriate issues/PR template. In addition, check for existing open issues and pull requests to ensure that someone else has not already corrected the information. If you need any help, feel free to tag [@AruneshSingh](https://github.com/AruneshSingh) in your discussions/issues/PRs. From 3424f49e39a4aa0eace6c13bf4e12d7d1ae91d5e Mon Sep 17 00:00:00 2001 From: svonava Date: Tue, 25 Aug 2026 12:54:51 -0700 Subject: [PATCH 2/2] Potential fix for pull request finding Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com> --- .../retrieval_augmented_generation_eval_qdrant_ragas.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md b/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md index 1d4761633..8b7b08b88 100644 --- a/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md +++ b/docs/articles/retrieval_augmented_generation_eval_qdrant_ragas.md @@ -251,7 +251,7 @@ EVAL_SIZE = 10 RETRIEVAL_SIZE_3 = 3 ## The dataset used to evaluate RAG using RAGAS -## Note this is the dataset needed for evaluation hence has to be recreated every time changes to RAG config is made +## Note: this is the dataset needed for evaluation, so it must be recreated every time changes to the RAG config are made rag_eval_dataset_512_3 = create_eval_dataset(qdrant_qna_dataset,EVAL_SIZE,RETRIEVAL_SIZE_3) # The dataset is then exported as a CSV file, with a filename that includes details of the experiment for easy identification, such as the chunk size along with retrieval window used in this case rag_response_dataset_512_3 = Dataset.from_dict(rag_eval_dataset_512_3)