Quick Start Guide

Prerequisites

Before using the Testing API, ensure you have the following installed:

• Python 3.8+
• pip (Python package manager)
• An API Key: Issued via the Supervised AI platform dashboard.

Installation

Clone the repository and install the necessary dependencies to interact with the testing environment:

git clone https://github.com/UditAkhourii/testing-api.git
cd testing-api
pip install -r requirements.txt

Authentication

All requests to the Supervised AI Testing API must include your API Key in the request header.

| Header | Type | Description | | :--- | :--- | :--- | | Authorization | string | Your API Key (e.g., Bearer YOUR_API_KEY) | | Content-Type | string | Must be application/json |

Quick Usage Example

The following Python script demonstrates how to initialize the client and perform a basic connectivity test to the Supervised AI platform.

import requests

# Configuration
API_URL = "https://api.supervised.ai/v1/test"
HEADERS = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

def check_connection():
    response = requests.get(f"{API_URL}/ping", headers=HEADERS)
    
    if response.status_code == 200:
        print("Success:", response.json())
    else:
        print(f"Error {response.status_code}: {response.text}")

if __name__ == "__main__":
    check_connection()

Core Endpoints

1. Health Check

Validates the availability of the testing infrastructure.

• Endpoint: /ping
• Method: GET
• Output:
  • status (string): Connection status (e.g., "pong").
  • timestamp (iso8601): Current server time.

2. Run Test Inference

Simulates a model inference call within the testing structure.

• Endpoint: /inference/test

• Method: POST

• Input (JSON): | Parameter | Type | Required | Description | | :--- | :--- | :--- | :--- | | model_id | string | Yes | The unique identifier of the AI model. | | input_data | dict | Yes | The payload/features to be processed. | | stream | boolean | No | Whether to stream the response (default: false). |

• Output (JSON):

  • request_id (uuid): Unique identifier for the test session.
  • prediction (object): The result generated by the model.
  • latency_ms (integer): Time taken to process the request.

Error Handling

The API uses standard HTTP response codes:

• 200 OK: Request succeeded.
• 401 Unauthorized: Invalid or missing API Key.
• 422 Unprocessable Entity: Validation error in your input JSON.
• 500 Internal Server Error: Issue within the Supervised AI testing infrastructure.