MLOps Batch Job

Run locally

pip install -r requirements.txt

python run.py --input data.csv --config config.yaml --output metrics.json --log-file run.log

Run with Docker

docker build -t mlops-task . docker run --rm mlops-task

Description

• Load config
• Validates dataset
• Computes rolling mean on close
• Generates binary signal
• Outputs metrics.json and run.log

Example metrics.json

Success

{ "version": "v1", "rows_processed": 10000, "metric": "signal_rate", "value": 0.4991, "latency_ms": 119, "seed": 42, "status": "success" }

Error

{ "version": "v1", "status": "error", "error_message": "Missing required column: close" }

Notes

• First window-1 rows produce NaNs and are excluded
• Metrics file is always written (success/error)