[WIP] Adding a A large EEG database with users' profile information for motor imagery Brain-Computer Interface dataset #404
Conversation
… into Dreyer2023_MI_dataset
moabb/datasets/Dreyer2023.py
Outdated
|
|
||
| subj_id = self.db_id + str(subject + self.db_idx_off[self.db_id]) | ||
|
|
||
| ch_names = [ |
There was a problem hiding this comment.
To avoid long list like that you could use #fmt: off/on like explained here
moabb/datasets/Dreyer2023.py
Outdated
| } | ||
|
|
||
|
|
||
| class Dreyer2023(BaseDataset): |
There was a problem hiding this comment.
You may rename this class as Dreyer2023Base to make clear that is the parent class and you could indicate in the docstring that it should not be used directly, instead use Dreyer2023A, Dreyer2023BorDreyer2023C`.
moabb/datasets/Dreyer2023.py
Outdated
| .. admonition:: Dataset summary | ||
|
|
||
| ========== ======= ======= ========== ================= ============ =============== =========== | ||
| Name #Subj #Chan #Classes #Trials / class Trials len Sampling rate #Sessions | ||
| ========== ======= ======= ========== ================= ============ =============== =========== | ||
| Dreyer2023 87 27 2 20 5s 512 Hz 6 | ||
| ========== ======= ======= ========== ================= ============ =============== =========== | ||
|
|
||
| =================== | ||
| Dataset description | ||
| =================== | ||
| A large EEG database with users' profile information for motor imagery | ||
| Brain-Computer Interface research | ||
|
|
||
| Data collectors : Appriou Aurélien; Caselli Damien; Benaroch Camille; | ||
| Yamamoto Sayu Maria; Roc Aline; Lotte Fabien; | ||
| Dreyer Pauline; Pillette Léa | ||
| Data manager : Dreyer Pauline | ||
| Project leader : Lotte Fabien | ||
| Project members : Rimbert Sébastien; Monseigne Thibaut |
There was a problem hiding this comment.
The dataset summary + description should removed here and put in each dataset (A, B and C) with the correct number of subjects. Keep only the information that this is the parent class and that it should not be instanciated directly.
There was a problem hiding this comment.
Thanks, @sylvchev ! I moved dataset summary and description to each A, B and C dataset. Since, recording procedure is the same, a large portion of the descriptions is the same for all the three datasets.
| from sklearn.pipeline import make_pipeline | ||
|
|
||
| import moabb | ||
| from moabb.datasets import Dreyer2023A, Dreyer2023B, Dreyer2023C |
There was a problem hiding this comment.
| from moabb.datasets import Dreyer2023A, Dreyer2023B, Dreyer2023C | |
| from moabb.datasets import Dreyer2023A |
B & C not used here
| from moabb.paradigms import MotorImagery | ||
|
|
||
|
|
||
| dreyer2023 = Dreyer2023A() |
There was a problem hiding this comment.
| dreyer2023 = Dreyer2023A() | |
| dreyer2023 = Dreyer2023A() | |
| dreyer2023.subject_list = dataset.subject_list[:6] |
Generating documentation uses lots of computational ressources, both for CPU and network bandwidth. We try to limit as much as possible the datasets downloaded for generating docs (they are shared between the examples). As the user's profile included in this dataset could give some interesting insight, we could consider to add this dataset in the documentation, but it should use only a selection of subjects to keep the bandwidth usage within reasonable limits. Could you adapt the above line to select only a representative individual for your analysis below. If you could keep the number of subjects between 4 and 8, it will really lower the doc building time.
Cherry picking subjects for the documentation is allowed ^^
There was a problem hiding this comment.
Thanks! I've selected 4 subjects (to include all combinations of participants' and experimenters' genders). However, since data of all subjects are zipped they all need to be downloaded together, is there some solution for this?
There was a problem hiding this comment.
No unfortunately.
The only possibility I see is to make this example a static one, that is you change the name of the file into Dreyer_clf_scores_vs_subj_info.py (without the plot_ prefix) and Sphinx won't execute this example, only display the code. If you want, you could generate a figure on your computer and upload it in a new folder examples/images/, then you could include the resulting image from your example code (see for example this page or that page).
| from moabb.paradigms import MotorImagery | ||
|
|
||
|
|
||
| dreyer2023 = Dreyer2023A() |
There was a problem hiding this comment.
No unfortunately.
The only possibility I see is to make this example a static one, that is you change the name of the file into Dreyer_clf_scores_vs_subj_info.py (without the plot_ prefix) and Sphinx won't execute this example, only display the code. If you want, you could generate a figure on your computer and upload it in a new folder examples/images/, then you could include the resulting image from your example code (see for example this page or that page).
|
Can you review @sylvchev? |
This PR aims to add a new MI dataset https://zenodo.org/record/7554429 proposed in #1 .