Datasets API
The Datasets API provides comprehensive dataset management capabilities for NER (Named Entity Recognition), text classification, and relation extraction tasks. It supports creating, listing, updating, deleting datasets, and uploading files in various formats (JSONL, CSV, Parquet, JSON).
Available Endpoints
- List Datasets - Retrieve a paginated list of datasets with optional filtering
- Create Dataset - Create a new dataset with optional file upload
- Get Dataset - Retrieve detailed information about a specific dataset
- Update Dataset - Update dataset metadata (name and/or description)
- Delete Dataset - Delete a dataset permanently
- Upload File - Upload a file to an existing dataset
- List Formats - Retrieve a list of available dataset formats
- Generate Dataset Plan - Generate a structured schema and sample plan from a prompt
- Generate Dataset - Trigger an asynchronous synthetic dataset generation pipeline
Getting Started
Most Datasets API endpoints require authentication:
| Header | Description |
|---|---|
KnowledgatorAPI-Key | Your API key |
Content-Type | application/json for JSON requests, multipart/form-data for file uploads |
Example Usage
SDK Usage
💡 Dedicated SDK Guides: See Python SDK | TypeScript SDK
TypeScript/JavaScript
import { createClient } from '@knowledgator/typescript-sdk';
const kg = createClient('YOUR_API_KEY');
// Use any of the datasets endpoints
const datasets = await kg.datasets.list();
Python
from knowledgator import Knowledgator
kg = Knowledgator(api_key="YOUR_API_KEY")
# Use any of the datasets endpoints
datasets = kg.datasets.list()
cURL
curl -X GET "https://api.platform.knowledgator.com/datasets/" \
-H "KnowledgatorAPI-Key: YOUR_API_KEY"
Dataset Types
ner– Named Entity Recognitiontext-classification– Text Classificationrelation-extraction– Relation Extraction
Dataset Formats
jsonl– JSON Lines (one JSON object per line)csv– Comma-Separated Valuesparquet– Apache Parquet columnar formatjson– Standard JSON format
Dataset Status
pending– Dataset created, awaiting processingprocessing– Dataset is being processedready– Dataset is ready for usefailed– Processing failed
For detailed documentation on each endpoint, click on the links above.