AE Server (Autoencoder-based Anomaly Detection)

Quickstart

1. Create env and install deps (uv 추천):

# using uv (project dir는 `ae_server/` 입니다)
cd /data/janghyunroh/projects/hackathon/anomaly_detection_server/ae_server
# (선택1) 자동으로 .venv 생성+설치: 프로젝트 의존성 싱크
uv sync
# (선택2) 수동: venv 만들고 에디터블 설치
uv venv
source .venv/bin/activate
uv pip install -e .

# CPU 전용 PyTorch가 필요하면(대용량 CUDA 미설치)
# uv pip install --index-url https://download.pytorch.org/whl/cpu torch==2.3.1

2. Prepare training data at /data/janghyunroh/projects/hackathon/anomaly_detection_server/data/normal_traffic.jsonl.

3. Train model and save artifacts:

uv run -m ae_server.app.training --epochs 5

4. Run server:

uv run -m uvicorn ae_server.app.main:app --host 0.0.0.0 --port 8080 --reload

5. Test request:

uv run /data/janghyunroh/projects/hackathon/anomaly_detection_server/scripts/test_client.py