nextpoints-sdk 0.2.0

Shared data models and service protocols for NextPoints and its ML backend

nextpoints-sdk

Shared data models and simple synchronous service protocol client between NextPoints and its ML backend.

MVP scope:

• Pydantic models for project metadata, annotations, calibration.
• Synchronous HTTP client (httpx) for three service endpoints: pre-annotation, reconstruction, calibration.
• Minimal error and artifact models.

Install (local)

pip install -e .

Run tests

pytest -q

Basic Usage

from nextpoints_sdk.client.services import ServicesClient
from nextpoints_sdk.models.requests import PreAnnotationRequest
# build project_meta ...
client = ServicesClient(base_url="http://ml-backend:8000", api_key="your-key")
resp = client.pre_annotate(PreAnnotationRequest(project_meta=project_meta))
print(resp.annotations.keys())

This is an early skeleton. Models will be filled incrementally.