Featured run
Current source: combined_20260319_081200_pruned_rest_events_0_1_2 · 20260319_075520
Action accuracy: 89.79%
Finger accuracy on non-rest windows: 85.96%
Event-level joint accuracy: 87.60%
Saved test split
89.79% / 85.96%
Action and finger accuracy
Action accuracy is reported on all test windows; finger accuracy is reported on non-rest windows.
2,301 test windows; 1,994 non-rest
Primary holdout
84.66%
Joint accuracy with rest and applicability diagnostics
Adds rest and applicability diagnostics beyond the saved test split.
Deployment pair invariant passed; 2.26% applicability FN on true non-rest
Event-level holdout
92.56% / 87.60%
Action and joint accuracy by event
Majority vote over all held-out windows belonging to the same session event.
121 total events; 118 non-rest events
Chronological replay
84.30%
Core full-session replay
Replay across the two core movement sessions on a longer chronological trace.
95.99% Rest TPR; 18.07% applicability FP on true rest
Pseudo-live replay
86.42%
Current per-finger decision path
Replay through the current decision path, including precision, event hit rate, and false rest actuation.
0.25% false REST actuation; 95.37% would-send precision; 31.56% throughput coverage
Harder replay session
71.96%
March 17 realism session
A harder pseudo-live check that exposes where applicability recall is still weak.
52.98% applicability FN on true non-rest
Selection context
The trained checkpoint is still 20260319_075520. The public deployment bundle has been updated around it: same-finger holds prevent command chatter, other fingers can still actuate immediately, and the public metrics now report throughput separately from accuracy.
95.37%
Would-send precision
Among non-rest windows that passed the current per-finger actuation gate, 95.37% carried the correct movement command.
0.25%
False rest actuation
Only 6 of 2,404 true REST replay windows produced a movement command under the current defaults.
91.11%
Event hit rate
The replay hit 543 of 596 movement events at least once, which is the better responsiveness summary than window-level send coverage.
2,595 configs
Postprocess ablation
The March 16, 2026 website update documents a 2,595-config ablation over thresholds, smoothing, hysteresis, adjacency, and finger-mode settings.
How the featured run was chosen
Training snapshot
Architecture
CNNLSTMFingerActionNet
The March 19 checkpoint combines action decoding, active-finger decoding, and a dedicated finger-applicability head.
Optimization
60 epochs · batch 64 · lr 0.001 · seed 43
These values come from the restored March 19 deployment metrics bundle.
Split policy
group_trial · test_size 0.2 · calibration_size 0.1
The holdout bundle stays tied to a fixed split while calibration is separated from the main train/test partition.
Input + preprocessing
64 x 4 windows · center_detrend
Per-window centering and detrending are frozen into the reference run's preprocessing and normalizer config.
Interactive latent space
Each point is one EEG window projected from the learned latent representation into a three-component PCA view. PCA is shown here because it preserves a linear comparison across the full dataset, the training split, and the held-out test split.
The train/test pair stays colored by true finger so the geometry is directly comparable across splits. Correctness and deployment-gating views are available on the visualization page.
PCA · Full dataset
Each point is one EEG window, projected into three principal components and colored by the labeled finger.
Separated regions suggest structured learned organization, while overlap marks similar or harder windows.
Train/test comparison at a glance
Full dataset: 12,447 windows. Train split: 10,146 windows. Held-out test split: 2,301 windows. Scan the pair below for whether the held-out geometry still resembles the training structure under the same color coding.
Same projection family, same label colors, different split membership.
PCA · Train split
This restricts the PCA view to the 10,146 training windows while keeping the same true-finger coloring used in the full-dataset view.
Keeping train and test on the same coloring makes split-to-split geometry easier to compare directly.
PCA · Test split
This shows the 2,301 held-out test windows only, using the same true-finger coloring so the class structure can be compared against the training split.
If the held-out view preserves similar neighborhoods rather than collapsing, the representation is carrying structure beyond the fitting set.
These figures show error structure and confidence behavior for the featured run.
Action confusion matrix
REST, OPEN, and CLOSE confusion for the featured bundle.
Finger confusion matrix (non-rest)
Finger-level confusion after rest windows are removed from the task.
Evaluation summary
Confusion matrices and reliability diagrams for the featured bundle.
Confidence and uncertainty scatter
Confidence spread for the featured bundle.
These figures add signal-distribution context for the decoder metrics above and help explain where the model separates movement states.
Action alpha rest-delta topomap
Rest-relative alpha maps for REST, OPEN, and CLOSE in the March 19 winning session. OPEN and CLOSE both show the dominant TP10 decrease and smaller TP9 increase that characterize the current 2-M16 action story.
Finger alpha rest-delta topomap with NONE reference
Finger-level rest-delta maps, including NONE as the explicit rest reference. The strongest variation remains concentrated on TP10 and TP9, which helps explain why lateral Muse 2 channels carry most of the finger-separation load.
The table keeps the earlier 1-M16 bundle visible as a methods baseline while separating it from the current 2-M16 deployment-facing stack.
2-m16
March 19, 2026
Featured run
Action accuracy
89.79%
Finger accuracy
85.96%
Test windows
2,301
1-m16-500
March 5, 2026
Historical baseline
Action accuracy
83.94%
Finger accuracy
80.61%
Test windows
2,652
| Run | Role | Action accuracy | Finger accuracy on non-rest | Test windows | Links |
|---|---|---|---|---|---|
2-m16 March 19, 2026 | Featured run | 89.79% | 85.96% | 2,301 | |
1-m16-500 March 5, 2026 | Historical baseline | 83.94% | 80.61% | 2,652 |
Deep-dive page
Tuning scale, model-selection logic, replay gates, and recommended context.
Featured run detail
The full training, tuning, replay, and selection story behind the current bundle.
Featured HTML report
The full figure and metrics report for the featured run.
Sanitized metrics JSON
Machine-readable metrics, figures, and benchmark fields for the current bundle.
May 6 per-finger actuation update
Current 2-M16 deployment metrics, event-hit reporting, and corrected hold semantics.
April 24 model-selection audit
The historical audit that restored March 19 before the per-finger replay update.
April 29 live-inference diagnosis
The same-sitting replay failure, signal-shift finding, and stricter preflight requirement.
March 16 split-fix update
The update that documents the 2,595-config postprocess ablation.
Older tuning update
The earlier update that documented 100+ trained variants and a 30+ hour sweep.
All dated updates
The chronological project trail behind the current metrics.
Research repo
Clone the project, inspect the artifacts, reproduce results, or contribute.
Poster and paper status
Public ISEF poster board PDF plus the current private manuscript status.