Tutorials: the neuralset pipeline

Each tutorial covers one step of the pipeline — from loading data through to building a DataLoader — with code you can run and modify.

Note

Concepts at a glance: Study → Events DataFrame → Transforms → Segmenter → Dataset → DataLoader

Everything stays lightweight (metadata only) until you call the DataLoader. Every step is cacheable via exca.

Study

Interface to an external dataset. Download, iterate timelines, load events.

Snippet Tutorial

study = ns.Study(name="Fake2025Meg",
                 path=ns.CACHE_FOLDER)
events = study.run()
print(f"{len(events)} events, "
      f"{events['subject'].nunique()} subjects")

Open full tutorial →

↓

Events DataFrame

Typed DataFrame rows. Neural, stimuli, text — everything is an event.

Snippet Tutorial

from neuralset.events import Event
evt = Event(type="Word", start=1.0,
            duration=0.3, timeline="sub-01")
print(evt)           # pydantic model
print(evt.model_dump())  # dict

Open full tutorial →

↓

Transforms

Modify the events DataFrame: split, chunk, align, add context.

Snippet Tutorial

import neuralset as ns
transform = ns.events.transforms.AddSentenceToWords()
events = transform(events)
print(events[events.type == "Sentence"].head())

Open full tutorial →

↓

Extractors

Convert events into tensors. EEG, fMRI, text, images, audio.

Snippet Tutorial

meg = ns.extractors.MegExtractor(frequency=100.0)
freq = ns.extractors.WordFrequency(language="english")
sample = meg(events, start=0.0, duration=1.0)
print(f"MEG shape: {sample.shape}")

Open full tutorial →

↓

Segmenter & Dataset

Time segments around triggers, then a torch.utils.data.Dataset ready for a DataLoader.

Snippet Tutorial

meg = ns.extractors.MegExtractor(frequency=100.0)
segmenter = ns.dataloader.Segmenter(
    start=-0.1, duration=0.5,
    trigger_query='type=="Word"',
    extractors=dict(meg=meg),
    drop_incomplete=True)
dataset = segmenter.apply(events)
loader = DataLoader(dataset, batch_size=8,
                    collate_fn=dataset.collate_fn)

Open full tutorial →

↓

Putting it Together

Compose full pipelines: Studies + Transforms + Extractors + Segmenter, all in one config.

Snippet Tutorial

import neuralset as ns
from neuralset.events import transforms
chain = ns.Chain(steps=[
    ns.Study(name="Fake2025Meg",
             path=ns.CACHE_FOLDER),
    transforms.AddSentenceToWords(),
])
events = chain.run()

Open full tutorial →

← Installation Tutorial 1: Events →

Events

Events

Studies

Studies

Transforms

Transforms

Extractors

Extractors

Segmenter & Dataset

Segmenter & Dataset

Chains

Chains