thalos-housekeeping-safety-v1

Thalos Housekeeping & Obstruction Safety – v1.0

(Roboflow → HuggingFace export)

This repository contains the Housekeeping & Obstruction Safety Detection model used in the Thalos Safety Intelligence pipeline.

It identifies housekeeping and obstruction-related hazards such as: • tripping hazards • misplaced pallets or boxes • walkway obstructions • storage blockages • cluttered work areas • general housekeeping violations in industrial, warehouse, and facility settings

This model was originally trained in Roboflow (Object Detection Large / YOLOv8-derived) and exported as a PyTorch .pt file. It is now hosted on HuggingFace for independent, cost-controlled inference inside the Thalos YOLO service.

⸻

🧹 Usage (Python)

from ultralytics import YOLO

model = YOLO("thalostech2025/thalos-housekeeping-safety-v1/housekeeping_obstructions_weights.pt")

results = model("example.jpg") results.show()

⸻

🌐 Load From URL

(used by Thalos YOLO Service)

import torch import requests from io import BytesIO

HF_URL = "https://huggingface.co/thalostech2025/thalos-housekeeping-safety-v1/resolve/main/housekeeping_obstructions_weights.pt"

response = requests.get(HF_URL) model = torch.load(BytesIO(response.content), map_location="cpu")

⸻

🏷️ Labels / Classes

The following hazard classes are included:

pallet box storage_block walkway_obstruction misc_item

(May include additional classes depending on v1.0 dataset structure.)

⸻

📄 License

This model is distributed under the AGPL-3.0 license, consistent with Roboflow export requirements.

⸻

⚠️ Important Notes • This repository only contains weights, not training data. • Thalos uses this model via server-side inference in the YOLO microservice. • Future versions (v2+) may be distilled or retrained for performance and speed improvements.

⸻

👤 Maintainer

Thalos Tech (2025) Safety Intelligence & Hazard Detection Platform

⸻