bruAristimunha commited on
Commit
4365ee8
·
verified ·
1 Parent(s): 1bad3fd

Complete training-data provenance and Braindecode software citation

Browse files

README-only metadata audit, 2026-11-09. Braindecode Zenodo citation. No weights, config, license, or upstream paper citations changed.

Files changed (1) hide show
  1. README.md +35 -0
README.md CHANGED
@@ -32,3 +32,38 @@ accessions, Hub ids otherwise). `nm`/`on` ids are datasets on [NEMAR](https://ne
32
  - Supervised SleepStagerChambon2018 (5 classes). SleepPhysionet subject_ids=[0, 1], recording_ids=[2] (second night only), crop_wake_mins=30 -> 2 recordings: index 0 = subject 0 night 2 (SC4002) trains, index 1 = subject 1 night 2 (SC4012) is the validation recording used by EarlyStopping(load_best=True) to select the epoch. EEG channels Fpz-Cz and Pz-Oz, 30 s windows at 100 Hz, sequences of 3 windows. Split is BaseConcatDataset.split(dict(train=..., valid=...)), whose values are RECORDING INDICES into the concatenated dataset, not subject ids (braindecode/datasets/base.py split: `[self.datasets[ds_ind] for ds_ind in ds_inds]`); SleepPhysionet orders recordings subject by subject, night 1 before night 2 (mne fetch_data loops `for subject in subjects`).
33
  - Sources: braindecode git 05e606dd^:scripts/train_and_push_tutorial_checkpoints.py (offline trainer that produced the tutorial checkpoints; removed from the repo in 05e606dd, 2026-04-19, and moved to https://gist.github.com/bru...; W/bd/examples/applied_examples/plot_sleep_staging_chambon2018.py; https://huggingface.co/braindecode/plot_sleep_staging_chambon2018/resolve/main/metadata.json.
34
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32
  - Supervised SleepStagerChambon2018 (5 classes). SleepPhysionet subject_ids=[0, 1], recording_ids=[2] (second night only), crop_wake_mins=30 -> 2 recordings: index 0 = subject 0 night 2 (SC4002) trains, index 1 = subject 1 night 2 (SC4012) is the validation recording used by EarlyStopping(load_best=True) to select the epoch. EEG channels Fpz-Cz and Pz-Oz, 30 s windows at 100 Hz, sequences of 3 windows. Split is BaseConcatDataset.split(dict(train=..., valid=...)), whose values are RECORDING INDICES into the concatenated dataset, not subject ids (braindecode/datasets/base.py split: `[self.datasets[ds_ind] for ds_ind in ds_inds]`); SleepPhysionet orders recordings subject by subject, night 1 before night 2 (mne fetch_data loops `for subject in subjects`).
33
  - Sources: braindecode git 05e606dd^:scripts/train_and_push_tutorial_checkpoints.py (offline trainer that produced the tutorial checkpoints; removed from the repo in 05e606dd, 2026-04-19, and moved to https://gist.github.com/bru...; W/bd/examples/applied_examples/plot_sleep_staging_chambon2018.py; https://huggingface.co/braindecode/plot_sleep_staging_chambon2018/resolve/main/metadata.json.
34
 
35
+
36
+
37
+ ## Cite Braindecode
38
+
39
+ If you use this braindecode port in scientific work, cite the software on Zenodo,
40
+ [doi:10.5281/zenodo.17699192](https://doi.org/10.5281/zenodo.17699192),
41
+ and the original model and datasets cited above.
42
+
43
+ ```bibtex
44
+ @software{braindecode,
45
+ author = {Aristimunha, Bruno and
46
+ Guetschel, Pierre and
47
+ Wimpff, Martin and
48
+ Gemein, Lukas and
49
+ Rommel, Cedric and
50
+ Banville, Hubert and
51
+ Sliwowski, Maciej and
52
+ Wilson, Daniel and
53
+ Brandt, Simon and
54
+ Gnassounou, Théo and
55
+ Paillard, Joseph and
56
+ {Junqueira Lopes}, Bruna and
57
+ Sedlar, Sara and
58
+ Moreau, Thomas and
59
+ Chevallier, Sylvain and
60
+ Gramfort, Alexandre and
61
+ Schirrmeister, Robin Tibor},
62
+ title = {Braindecode: toolbox for decoding raw electrophysiological brain data
63
+ with deep learning models},
64
+ url = {https://github.com/braindecode/braindecode},
65
+ doi = {10.5281/zenodo.17699192},
66
+ publisher = {Zenodo},
67
+ license = {BSD-3-Clause},
68
+ }
69
+ ```