Self Learning Postgres DB

Pricing

from $0.30 / 1,000 results

Self Learning Postgres DB

Self-learning vector database with GNN-powered index optimization. Features: vector search, RAG queries, embeddings, clustering, deduplication, batch ops, and data import/export. Scales with Raft consensus.

Pricing

from $0.30 / 1,000 results

Rating

0.0

(0)

Developer

Reuven Cohen

Maintained by Community

Actor stats

Bookmarked

Total users

Monthly active users

3 days ago

Last modified

Action

actionEnumRequired

The operation to perform on the vector database

Value options:

"full_workflow": string"search": string"insert": string"batch_insert": string"get": string"list": string"update": string"delete": string"upsert": string"hybrid_search": string"multi_query_search": string"mmr_search": string"graph_search": string"range_search": string"batch_search": string"create_table": string"drop_table": string"list_tables": string"table_stats": string"create_index": string"reindex": string"train_gnn": string"optimize_index": string"analyze_patterns": string"sona_learn": string"sona_status": string"cluster": string"find_duplicates": string"deduplicate": string"export": string"import": string"rag_query": string"summarize": string"ping": string"version": string"embedding_models": string"generate_embedding": string"similarity": string

Default value of this property is "full_workflow"

Database Connection

connectionStringstringOptional

PostgreSQL connection URL. Leave empty for embedded database (non-persistent). For persistent storage, use your own PostgreSQL with ruvector/pgvector extension.

Table/Collection Name

tableNamestringOptional

Name of the vector table (collection)

Default value of this property is "documents"

Search Query

querystringOptional

Natural language query for semantic search. The AI understands meaning, not just keywords.

Query Vector

queryVectorarrayOptional

Pre-computed embedding vector (alternative to query text). Use with external embedding APIs.

Documents

documentsarrayOptional

Documents to insert. Each should have 'content' and optional 'metadata' and 'embedding'.

Number of Results

topKintegerOptional

Maximum number of results to return

Default value of this property is 10

Distance Metric

distanceMetricEnumOptional

How to measure vector similarity

Value options:

"cosine": string"l2": string"inner_product": string"manhattan": string

Default value of this property is "cosine"

Filter

filterstringOptional

SQL WHERE clause for filtering. Example: metadata->>'category' = 'AI'

Minimum Score

minScorenumberOptional

Minimum similarity score threshold (0-1)

Default value of this property is 0

Maximum Distance

maxDistancenumberOptional

Maximum distance threshold for range search

Include Embeddings

includeEmbeddingsbooleanOptional

Include embedding vectors in results (increases response size)

Default value of this property is false

Include Metadata

includeMetadatabooleanOptional

Include metadata in results

Default value of this property is true

Embedding Model

embeddingModelEnumOptional

AI model for generating text embeddings. No API key needed - runs locally!

Value options:

"all-MiniLM-L6-v2": string"bge-small-en-v1.5": string"bge-base-en-v1.5": string"nomic-embed-text-v1": string"gte-small": string"e5-small-v2": string

Default value of this property is "all-MiniLM-L6-v2"

Generate Embeddings

generateEmbeddingsbooleanOptional

Auto-generate embeddings for documents without them

Default value of this property is true

Vector Dimensions

dimensionsintegerOptional

Embedding dimensions (384 for MiniLM/BGE-small, 768 for larger models)

Default value of this property is 384

Index Type

indexTypeEnumOptional

Vector index algorithm for faster search

Value options:

"hnsw": string"ivfflat": string"none": string

Default value of this property is "hnsw"

HNSW M Parameter

hnswMintegerOptional

Max connections per node. Higher = better recall, more memory

Default value of this property is 16

HNSW ef_construction

hnswEfConstructionintegerOptional

Index build quality. Higher = better index, slower build

Default value of this property is 64

HNSW ef_search

hnswEfSearchintegerOptional

Search quality. Higher = better recall, slower search

Default value of this property is 100

IVF Lists

ivfListsintegerOptional

Number of IVF partitions for IVFFlat index

Default value of this property is 100

Hybrid Weight

hybridWeightnumberOptional

Balance between vector (1.0) and keyword (0.0) search

Default value of this property is 0.7

Batch Size

batchSizeintegerOptional

Documents per batch for batch operations

Default value of this property is 100

Document ID

documentIdintegerOptional

Single document ID for get/update/delete operations

Document IDs

documentIdsarrayOptional

Multiple document IDs for batch delete

Updates

updatesobjectOptional

Fields to update: {content, metadata, embedding}

Enable Learning

enableLearningbooleanOptional

Enable self-learning index optimization

Default value of this property is false

Learning Rate

learningRatenumberOptional

GNN training learning rate

Default value of this property is 0.01

GNN Layers

gnnLayersintegerOptional

Number of Graph Neural Network layers

Default value of this property is 2

Training Epochs

trainEpochsintegerOptional

Number of GNN training epochs

Default value of this property is 10

Number of Clusters

numClustersintegerOptional

K-means cluster count

Default value of this property is 10

Clustering Algorithm

clusteringAlgorithmEnumOptional

Clustering method

Value options:

"kmeans": string"hierarchical": string

Default value of this property is "kmeans"

Similarity Threshold

similarityThresholdnumberOptional

Threshold for duplicate detection (0-1, higher = stricter)

Default value of this property is 0.95

Export Format

exportFormatEnumOptional

Data export format

Value options:

"json": string"csv": string

Default value of this property is "json"

Import Data

importDataarrayOptional

Data to import (array of documents with content and optional metadata)

RAG Max Tokens

ragMaxTokensintegerOptional

Maximum context tokens for RAG query

Default value of this property is 2000

RAG Context

ragContextstringOptional

Additional context to prepend to RAG results

Enable SONA Learning

sonaEnabledbooleanOptional

Enable TRM/SONA self-learning with trajectory tracking and pattern recognition

Default value of this property is true

EWC Lambda

ewcLambdanumberOptional

Elastic Weight Consolidation strength for anti-forgetting protection. Higher values preserve more learned knowledge.

Default value of this property is 2000

Pattern Threshold

patternThresholdnumberOptional

Minimum confidence threshold for pattern recognition (0-1)

Default value of this property is 0.7

Max Trajectories

maxTrajectoriesintegerOptional

Maximum number of trajectory steps to track for learning

Default value of this property is 100

Learning Tiers

sonaLearningTiersarrayOptional

SONA learning tiers to enable (instant=real-time, background=async, deep=comprehensive)

Default value of this property is ["instant","background"]

Supabase MCP SelfHosted

coupaul/Supabase-MCP-SelfHosted

🔒 Self-Hosted Supabase MCP Server - Enhanced Security Production-ready MCP server for self-hosted Supabase with enterprise security. Features SQL injection prevention, rate limiting, Docker optimization & direct DB access. Tools: database ops, auth, storage, types. Perfect for production!

Paul Roelens

Dealroom Scraper

saswave/dealroom-scraper

Dealroom company scraper. Collect data from companies using their url / domain name. Dealroom web crawler extract data about social networks, similarweb website traffic, funding rounds, investments, team members, KPI revenues and more ..

SASWAVE

106

Reddit Search (no login, unlimited, 9$)

maged120/reddit-search

the fastest most afordable reddit search

Maged

5.0

(1)

🔥 Glassdoor jobs scraper

bebity/glassdoor-jobs-scraper

ℹ️ Elevate your job search with the sleek Glassdoor Scraper for Apify. Tailor-made for the modern job seeker, it's your secret weapon in the job market. Dive into a world of opportunities with custom searches, precise filtering. Fast, efficient, and incredibly user-friendly 🚀✨

Bebity

1.6K

3.2

(6)

GLASSDOOR 🪟 Scraper HD

jupri/glassdoor

💫 All-in-One GlassDoor.com Scraper

cat

Advanced Social Media Agent

fiery_dream/advanced-social-media-agent

A production-grade AI agent built on cutting-edge research patterns for intelligent social media analysis on the Apify platform.

Cody Churchwell

Pdf Text Extractor Pro

dainty_screw/pdf-text-extractor-pro

PDF Text Extractor lets you quickly extract text from PDF files with high accuracy. Supports text chunking for AI, chatbots, and large language models (LLMs), making PDF-to-text conversion fast, clean, and ready for NLP or machine learning.

codemaster devops

5.0

(1)

Ai Question Generator

sunvic567/ai-question-generator

Generate educational questions from any document automatically. AI-powered quiz maker creates multiple choice, true/false, essay questions from PDFs and DOCX files. Perfect for teachers, trainers, and e-learning. Customize difficulty, export to JSON/CSV/Moodle XML. Turn documents into assessments.

Sunday Victor

4.0

(1)

Facebook Groups Scraper

apify/facebook-groups-scraper

Extract data from one or multiple public Facebook groups. Get group and post URLs, post text, comments, timestamp, likes and comments count, and basic commentator info. Download the data in JSON, CSV, and Excel and use it in apps, spreadsheets, and reports.

Apify

18K

4.3

(33)

Linkedin Sales Navigator Scraper

curious_coder/linkedin-sales-navigator-search-scraper

Scrape Linkedin sales navigator search results, lists, new results, profiles to extract social media urls, company website, positions

Curious Coder

6.1K

3.4

(25)

Linkedin Sales Navigator Scraper ( No cookies required )

muhammad_usama/Apify-Sales-Navigator-No-Cookies

Scrapes Linkedin sales navigator data ( People and Companies ).

Muhammad Usama

162

2.8

(3)

Self Learning Postgres DB

Self Learning Postgres DB

Action

Value options:

Database Connection

Table/Collection Name

Search Query

Query Vector

Documents

Number of Results

Distance Metric

Value options:

Filter

Minimum Score

Maximum Distance

Include Embeddings

Include Metadata

Embedding Model

Value options:

Generate Embeddings

Vector Dimensions

Index Type

Value options:

HNSW M Parameter

HNSW ef_construction

HNSW ef_search

IVF Lists

Hybrid Weight

Batch Size

Document ID

Document IDs

Updates

Enable Learning

Learning Rate

GNN Layers

Training Epochs

Number of Clusters

Clustering Algorithm

Value options:

Similarity Threshold

Export Format

Value options:

Import Data

RAG Max Tokens

RAG Context

Enable SONA Learning

EWC Lambda

Pattern Threshold

Max Trajectories

Learning Tiers

You might also like

Supabase MCP SelfHosted

Dealroom Scraper

Reddit Search (no login, unlimited, 9$)

🔥 Glassdoor jobs scraper

GLASSDOOR 🪟 Scraper HD

Advanced Social Media Agent

Pdf Text Extractor Pro

Ai Question Generator

Facebook Groups Scraper

Linkedin Sales Navigator Scraper

Linkedin Sales Navigator Scraper ( No cookies required )

Action

Value options:

Database Connection

Table/Collection Name

Search Query

Query Vector

Documents

Number of Results

Distance Metric

Value options:

Filter

Minimum Score

Maximum Distance

Include Embeddings

Include Metadata

Embedding Model

Value options:

Generate Embeddings