Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.nimbusbci.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Quickstart

Choose the SDK that matches your workflow. The Python SDK runs fully locally with an sklearn-style API, while the Julia SDK uses NimbusSDK.jl and requires an API key for the commercial core.

Python SDK Quickstart

Install nimbus-bci, fit your first classifier, and try batch, streaming, and active-learning workflows.

Julia SDK Quickstart

Install NimbusSDK.jl, authenticate with an API key, and run your first Julia inference workflow.

​
Before You Start

  • Use Python 3.11+ for the Python SDK or Julia 1.9+ for the Julia SDK.
  • Provide preprocessed EEG features, not raw EEG.
  • Start with the Preprocessing Requirements guide if you are not sure how to prepare CSP, bandpower, or ERP features.
Nimbus focuses on Bayesian inference and BCI workflow tooling. Filtering, artifact removal, epoching, and feature extraction happen before data reaches the SDK.

​
Common Next Steps

Model Selection

Compare NimbusLDA, NimbusQDA, NimbusSoftmax, NimbusProbit, and NimbusSTS.

Feature Normalization

Normalize features consistently across sessions.

Batch Processing

Evaluate complete trials offline and compute diagnostics.

Streaming Inference

Configure chunk-by-chunk low-latency inference.

​
FAQ

No. The Python SDK (nimbus-bci) runs locally and does not require an API key. The Julia SDK (NimbusSDK.jl) requires an API key to install and use NimbusSDKCore.
Choose Python if you want sklearn/MNE integration and local installation. Choose Julia if you want the NimbusSDK.jl workflow, RxInfer-backed models, or access to Julia model registry tooling.
Use extracted features, not raw EEG. Python examples usually use (n_trials, n_features) arrays for classifiers. Julia batch data usually uses (n_features, n_samples, n_trials) inside BCIData.