Basic Usage

This guide covers saving documents to your Endee vector store and querying them for similarity search.

Saving Documents

Store documents with metadata in your Endee vector store:


documents = [
    ("Python is dynamically typed.", {"creator": "Guido van Rossum", "typing": "dynamic"}),
    ("Go is statically typed.", {"creator": "Robert Griesemer", "typing": "static"})
]
 
for text, meta in documents:
    memory_store.save(text, meta)
 
print(f"Saved {len(documents)} documents")

Each document consists of:

Text content — The actual text to be embedded and stored
Metadata — Key-value pairs for filtering and organization

Querying the Vector Store

Perform semantic search to find relevant documents:


results = memory_store.search(
    query="Python typing discipline",
    limit=3,
    filter={"typing": {"$eq": "dynamic"}}
)
 
for r in results:
    print(f"ID: {r['id']}, Score: {r['score']}, Text: {r['context']}")

Query Parameters

Parameter	Type	Description
`query`	`str`	The search query text
`limit`	`int`	Maximum number of results to return
`filter`	`dict`	Optional metadata filter conditions

Filter Operators

Operator	Description	Example
`$eq`	Equal to	`{"typing": {"$eq": "dynamic"}}`
`$in`	In list	`{"tags": {"$in": ["ai", "ml"]}}`
`$range`	Numeric range	`{"score": {"$range": [70, 95]}}`

Query Results

Each result in the response contains:

Field	Description
`id`	Unique identifier for the document
`score`	Similarity score (higher is more relevant)
`context`	The original text content
`metadata`	Associated metadata

Complete Example


from endee_crewai import EndeeVectorStore
import time
 
# Initialize
embedder_config = {
    "provider": "cohere",
    "config": {"model_name": "small", "api_key": "<COHERE_API_KEY>"}
}
 
memory_store = EndeeVectorStore(
    type="programming_languages",
    api_token="<ENDEE_API_TOKEN>",
    embedder_config=embedder_config,
    space_type="cosine",
    crew=None,
)
 
# Save documents
documents = [
    ("Python is dynamically typed and great for beginners.", {"language": "python", "typing": "dynamic"}),
    ("Go is statically typed with built-in concurrency.", {"language": "go", "typing": "static"}),
    ("JavaScript runs in browsers and Node.js.", {"language": "javascript", "typing": "dynamic"}),
    ("Rust provides memory safety without garbage collection.", {"language": "rust", "typing": "static"})
]
 
for text, meta in documents:
    memory_store.save(text, meta)
 
time.sleep(1)  # Allow indexing
 
# Query for dynamic languages
results = memory_store.search(
    query="beginner friendly programming",
    limit=2,
    filter={"typing": {"$eq": "dynamic"}}
)
 
print("Dynamic languages for beginners:")
for r in results:
    print(f"  - {r['context']} (score: {r['score']:.4f})")