nemar-py 0.1.0

A Python client for accessing NEMAR datasets.

nemar-py

A Python client for accessing public NEMAR datasets through the NEMAR data endpoint.

Usage

nemar-py download --dataset nm000132 --target-dir data/nm000132

The downloader starts from:

https://data.nemar.org/{dataset}/

It uses the advertised dataset index, metadata URL, and version manifest. It does not use authentication and does not fall back to S3.

To inspect the versions already advertised by NEMAR:

nemar-py versions --dataset nm000132

--tag latest resolves through the endpoint's latest field, and explicit tags are validated against the endpoint's advertised versions.

By default, nemar-py uses aria2c when it is available on PATH; otherwise it uses a built-in HTTP downloader.

nemar-py download --dataset nm000132 \
                  --tag v1.1.1 \
                  --subject 001 \
                  --task MMN \
                  --datatype eeg \
                  --suffix eeg \
                  --extension .set

Semantic BIDS filters can be repeated. They accept either BIDS-prefixed labels (sub-001, task-MMN, run-01) or plain labels (001, MMN, 01). Less common BIDS entities can be selected with --entity key=value, for example --entity desc=clean.

Dataset-level BIDS folders are selected with --scope:

# Download stimuli for one task
nemar-py download --dataset nm000132 --scope stimuli --task MMN

# Download derivatives from one pipeline
nemar-py download --dataset nm000132 \
                  --scope derivatives \
                  --pipeline eeglab \
                  --subject 001

--include and --exclude remain available as path-level refinements. When combined with semantic BIDS filters, --include narrows the semantic result and --exclude removes paths from it.

Essential BIDS root files such as dataset_description.json, participants.tsv, participants.json, README, README.md, CHANGES, and LICENSE are kept when present, even when include or exclude patterns are used.

Python

import nemar

index = nemar.fetch_dataset_index(dataset="nm000132")
versions = nemar.list_dataset_versions(dataset="nm000132")

nemar.download(
    dataset="nm000132",
    target_dir="data/nm000132",
    subject=["001", "002"],
    task="MMN",
    datatype="eeg",
    suffix="eeg",
    extension=".set",
)

nemar.download(
    dataset="nm000132",
    scope="stimuli",
    task="MMN",
)