Skip to content

Download from NSRR

Download files from a dataset on the National Sleep Research Resource (NSRR).

Synopsis

somnio download-nsrr SLUG OUTPUT_DIR [OPTIONS]

See CLI for the full reference.

Before you start

Get an NSRR token

  • Create an account at sleepdata.org.
  • Request access to the dataset (e.g. SOF, SHHS, MESA).
  • After approval, copy your auth token from the dataset page or account profile.
  • Pass it with --token or set the NSRR_TOKEN environment variable (e.g. in .env).

Output layout

Files are placed under OUTPUT_DIR/SLUG/. If you use --path, only that subpath is created. For example:

  • somnio download-nsrr sof ./data ... → ./data/sof/...
  • somnio download-nsrr shhs ./data --path polysomnography ... → ./data/shhs/polysomnography/...

Examples

Full dataset

Download the full SOF dataset into ./data:

somnio download-nsrr sof ./data --token YOUR_NSRR_TOKEN

Single subpath

Download only the polysomnography folder from SHHS:

somnio download-nsrr shhs ./data --path polysomnography --token YOUR_NSRR_TOKEN

Using the environment for the token

Avoid putting the token on the command line by setting NSRR_TOKEN (e.g. in .env or your shell):

export NSRR_TOKEN=your_token_here
somnio download-nsrr sof ./data

Longer timeout and file logging

Use a 10-minute timeout and write logs to the default logs/ file:

somnio -l download-nsrr mesa ./datasets --token YOUR_NSRR_TOKEN --timeout-seconds 600

Custom log file

Send debug logs to a specific file for later inspection:

somnio --log-file ./nsrr-debug.log download-nsrr shhs ./data --token YOUR_NSRR_TOKEN

Retries and timeouts

For flaky connections or slow networks, use --timeout-seconds to increase the per-request timeout. You can also tune --download-retries (retries per file on connection/read timeout) and --http-retries (retries for API requests). See the CLI reference for defaults and details.