skills.rest
hung-phan
hung-phanCOMMUNITY
Explore all skills by this author →
GitHub

embeddings

Map unstructured text and image data to dense vectors for retrieval, clustering, and classification.

1|Updated Jun 4, 2026
Search Tags:#retrieval#classification#embedding#clustering#vector space
One-click install
npx skills add https://github.com/hung-phan/ml-skills --skill embeddings-hung-phan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: embeddings
Source: https://github.com/hung-phan/ml-skills/tree/main/skills/ml-review/references/ml-architectures/embeddings
Command: npx skills add https://github.com/hung-phan/ml-skills --skill embeddings-hung-phan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides tools for mapping unstructured data, like text and images, to dense vectors, enabling efficient retrieval, clustering, and classification tasks.

Core Features & Use Cases

  • Embedding Models: Offers various embedding models for text and images, including sentence-transformers, BGE, E5, and OpenAI/Voyage APIs.
  • Retrieval: Find the most relevant documents for a query in milliseconds over millions of vectors.
  • Clustering: Group similar items without labels.
  • Classification: Use vectors as features for downstream models.
  • Deduplication: Detect near-duplicates via cosine threshold.
  • Cross-modal Search: Match text queries to images (CLIP/SigLIP).
  • Use Case: Use this Skill to implement a semantic search engine that matches text queries to relevant images or to classify documents into categories based on their content.

Quick Start

Use the embeddings skill to generate embeddings for the text 'This is an example text'.

Frequently Asked Questions about embeddings

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I map unstructured text and images to dense vectors for semantic search?▼

To generate embeddings for unstructured data, this Skill maps text and images to dense vectors using models like sentence-transformers, BGE, E5, and OpenAI/Voyage APIs for retrieval tasks.

Can I use embeddings to match text queries directly to relevant images?▼

Yes, you can perform cross-modal search by matching text queries to images using embeddings generated via CLIP and SigLIP models for cross-modal retrieval.

What is the best way to group similar documents together without labels?▼

To group similar items without labels, this Skill generates dense vectors from unstructured data to perform efficient clustering of text and image datasets.

Does this Skill support detecting near-duplicate documents in a large dataset?▼

Yes, this Skill supports detecting near-duplicates by comparing dense vectors via a cosine similarity threshold, allowing efficient deduplication of unstructured text and image datasets.

How do I use dense vectors as features for downstream text classification models?▼

You can use dense vectors as features for downstream classification. This Skill maps unstructured text to embeddings, providing numerical representations required for text classification tasks.

Related Skills

Explore More Data & Analytics Skills→
lucifertrjlucifertrj
17

nltk-nlp-text-cleaning

Clean, tokenize, and normalize English text with NLTK for NLP pipelines.

Community
Advanced
zechenzhangAGIzechenzhangAGI
11.5k

faiss

Perform billion-scale k-NN similarity search with FAISS index types and GPU acceleration.

Community
Intermediate
ovachieverovachiever
52

aeon

Provide scikit-learn compatible time series machine learning algorithms for classification, regression, clustering, forecasting, and anomaly detection.

Community
Advanced
asvskartheekasvskartheek

tfidf-search

Build a TF-IDF search index to rank CSV text documents by cosine similarity.

Community
Intermediate