bruAristimunha commited on
Commit
162f5bc
·
verified ·
1 Parent(s): d716c94

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 +18 -0
README.md CHANGED
@@ -55,6 +55,24 @@ is fixed to 15 patches, so keep `n_times=22500`; you may freely change `n_chans`
55
  (channels are pooled) and `n_outputs` (the classification head is task-specific
56
  and randomly initialized — fine-tune it on your downstream task).
57
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  ## Usage
59
 
60
  ```python
 
55
  (channels are pooled) and `n_outputs` (the classification head is task-specific
56
  and randomly initialized — fine-tune it on your downstream task).
57
 
58
+ <!-- braindecode:training-data -->
59
+ ## Training data
60
+
61
+ The data that shaped these weights. The ids are the `datasets:` field of this card,
62
+ which braindecode reads into `model.card_data.datasets` when you call
63
+ `from_pretrained` (MOABB class names for MOABB datasets, OpenNeuro / NEMAR
64
+ accessions, Hub ids otherwise). `nm`/`on` ids are datasets on [NEMAR](https://nemar.org)
65
+ (`on` = NEMAR's copy of the OpenNeuro dataset with the same number). Rows marked
66
+ `private:` are data the authors did not release; they are not in `datasets:`.
67
+
68
+ | `datasets:` id | Dataset | Stage | Modality |
69
+ |---|---|---|---|
70
+ | `private:Brant-SEEG` | [Brant private SEEG corpus collected by the authors during routine epilepsy treatment (not released)](https://github.com/yzz673/Brant) | pretraining | iEEG |
71
+
72
+ - masked-autoencoding pretraining on the authors' private clinical SEEG corpus: 9 subjects, 2,528 h at 1000 Hz (1.01 TB), one hospital (unnamed); the separate 43 h private seizure set is downstream only and not in this checkpoint
73
+ - Sources: https://github.com/yzz673/Brant README; Zhang et al., NeurIPS 2023 (proceedings PDF via web.archive.org) p.4, Section 3.1.
74
+
75
+
76
  ## Usage
77
 
78
  ```python