Skip to content

accessing_results.md

Latest commit

 

History

History
77 lines (51 loc) · 2.3 KB

accessing_results.md

File metadata and controls

77 lines (51 loc) · 2.3 KB
Raw

Accessing Results

This page describes where VideoAnnotator stores outputs and how to retrieve them using the CLI or HTTP API.

For analysis, export everything as one table (CSV or Parquet), as ELAN files, or as a methods paragraph: see Results out.

Where results are stored

Docker Compose (recommended)

The default docker-compose.yml mounts these host folders into the container:

  • ./data → /app/data (input videos)
  • ./output → /app/output (job artifacts and exported results)
  • ./logs → /app/logs (server and job logs)
  • ./database → /app/database (SQLite database)

In practice:

  • Look for generated files under ./output/ on your host machine.
  • If you need to inspect container paths directly: docker compose exec videoannotator ls -la /app/output.

Local install

When running locally, the output directory is controlled by your configuration and/or storage settings. If you are using the default local setup, results typically land under ./output/ or ./storage/ depending on configuration.

Retrieve results with the CLI

All examples assume the API server is running on http://localhost:18011.

Check job status

uv run videoannotator job status <job_id>

If the job is still pending, the status response may include a queue_position field. queue_position is 1-based and indicates the job's current position among pending jobs (FIFO).

Get job results (JSON)

uv run videoannotator job results <job_id>

Download all job artifacts (ZIP)

uv run videoannotator job download-annotations <job_id> --output ./results/

If you are running via Docker Compose

docker compose exec videoannotator va job results <job_id>
docker compose exec videoannotator va job download-annotations <job_id> --output /app/output

Retrieve results with the HTTP API

export API_KEY="va_api_your_key_here"

# Get job status
curl -H "Authorization: Bearer $API_KEY" \
	"http://localhost:18011/api/v1/jobs/{job_id}"

# Get job results
curl -H "Authorization: Bearer $API_KEY" \
	"http://localhost:18011/api/v1/jobs/{job_id}/results"

# Download all artifacts (ZIP)
curl -H "Authorization: Bearer $API_KEY" \
	"http://localhost:18011/api/v1/jobs/{job_id}/artifacts" \
	--output "job_{job_id}_artifacts.zip"