bruAristimunha commited on
Commit
2701908
·
verified ·
1 Parent(s): 191f221

Card: list the training data (datasets:, base_model:) and add a Training data section

Browse files

Follow-up to braindecode PR #1493: `from_pretrained` reads `datasets:` into `model.card_data`. Ids: MOABB names, NEMAR nm/on ids, OpenNeuro ds ids; evidence per checkpoint.

Files changed (1) hide show
  1. README.md +22 -1
README.md CHANGED
@@ -6,9 +6,30 @@ tags:
6
  - braindecode
7
  - model_hub_mixin
8
  - pytorch_model_hub_mixin
 
 
9
  ---
10
 
11
  This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
12
  - Code: https://braindecode.org
13
  - Paper: [More Information Needed]
14
- - Docs: https://braindecode.org/stable/generated/braindecode.models.BENDR.html
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
  - braindecode
7
  - model_hub_mixin
8
  - pytorch_model_hub_mixin
9
+ datasets:
10
+ - TUEG
11
  ---
12
 
13
  This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
14
  - Code: https://braindecode.org
15
  - Paper: [More Information Needed]
16
+ - Docs: https://braindecode.org/stable/generated/braindecode.models.BENDR.html
17
+
18
+ <!-- braindecode:training-data -->
19
+ ## Training data
20
+
21
+ The data that shaped these weights. The ids are the `datasets:` field of this card,
22
+ which braindecode reads into `model.card_data.datasets` when you call
23
+ `from_pretrained` (MOABB class names for MOABB datasets, OpenNeuro / NEMAR
24
+ accessions, Hub ids otherwise). `nm`/`on` ids are datasets on [NEMAR](https://nemar.org)
25
+ (`on` = NEMAR's copy of the OpenNeuro dataset with the same number). Rows marked
26
+ `private:` are data the authors did not release; they are not in `datasets:`.
27
+
28
+ | `datasets:` id | Dataset | Stage | Modality |
29
+ |---|---|---|---|
30
+ | `TUEG` | [Temple University Hospital EEG Corpus (TUH EEG)](https://isip.piconepress.com/projects/nedc/html/tuh_eeg/) | pretraining | EEG |
31
+
32
+ - self-supervised (contrastive masked-sequence) on TUEG v1.1 and v1.2 (~1.5 TB EDF before preprocessing); no fine-tuning in this checkpoint (classification head randomly initialised)
33
+ - `TUEG` (the TUH EEG Corpus) contains the recordings of its sub-corpora TUAB, TUEV, TUSZ, TUAR, TUSL and TUEP; an evaluation on one of those is not held out from these weights unless the authors removed it.
34
+ - Sources: arXiv:2101.12037 p.5 (Sec. 3 / pre-training data); https://github.com/SPOClab-ca/BENDR/releases/tag/v0.1-alpha; HF commit braindecode/braindecode-bendr 2026-04-19 'Re-upload checkpoint in flat state_dict layout.'; size check.
35
+