Linux API & Runtime

The AugeLab Linux API provides a high-performance Python environment for executing industrial computer vision projects (.pmod) on Linux-based systems, edge devices, and embedded hardware like NVIDIA Jetson and Raspberry Pi.

Quick Start

The fastest way to get started is by installing the AugeLab Studio package via pip:

pip install studio
studio run your_project.pmod

Installation

Environment Requirements

AugeLab Runtime requires **Python 3.12**. We recommend using the **uv** package manager for optimal performance and dependency management.

pip install uv
uv venv --python 3.12 venv312
source venv312/bin/activate

Private Repository Configuration

To access AugeLab extra modules and CUDA-accelerated binaries, configure your pip settings:

[global]
index-url = https://pyrepo.augelab.com
extra-index-url = https://pypi.org
trusted-host = pyrepo.augelab.com

Docker Deployment

For containerized environments, use our optimized CUDA base images:

FROM public.ecr.aws/o9c1t9b5/augelab-studio/ubuntu-cuda12.8-py3.12-opencv-gpu:latest

RUN apt-get update && apt-get install -y --no-install-recommends \
    ca-certificates wget gnupg curl

# Install AugeLab Runtime
RUN uv pip install studio==3.4.5 --system --extra-index-url https://pyrepo.augelab.com

Python API Reference

Use the StudioScenario class to integrate AugeLab projects into your own Python applications:

from studio import StudioScenario

# Initialize with your verification code
scenario = StudioScenario(verification_code="YOUR_VERIFICATION_CODE")

# Load your .pmod project file
scenario.load_scenario("path/to/your_project.pmod")

# Main inference loop
while True:
    results = scenario.run(())
    if results[0]: # Exit condition if defined in project
        break

# Always cleanup resources after use
scenario.cleanup()

Note: The first argument of scenario.run()is always a tuple, used for passing dynamic parameters to the scenario.

Troubleshooting

If you encounter issues with camera access or performance:

Ensure your user has elevated rights (e.g., sudo) for USB camera access.
Verify CUDA compatibility with nvidia-smi.
Check logs in ~/.local/share/AugeLab Studio/logs.