You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Accessing/downloading datasets for YASA tutorials would be much easier with Pooch.
Add data files to OSF or Zenodo repository and remove from GitHub.
Create a datasets.py module to access downloaders
Update tutorials and notebooks with new way to load data
Pooch has very simple and clear documentation for this, and a great resource to work off of is Ensaio.
I'm afraid that some version of this will need to be implemented in v0.7. It feels necessary for me to make the tutorial notebooks for the new evaluation module, since those will require larger datasets to show how their purpose. (therefore kind of blocking #166)
**In the future it would be cool to access larger datasets here, like for full analyses rather than demos. It doesn't require any storage. A major benefit would be if YASA performed some standardization step, handling the various formats that people share their PSG data in. But given all the complexities that come with that, this feature should be reserved for a dedicated contributor later on (if interested in implementation at all).
The text was updated successfully, but these errors were encountered:
Noting here that we will use Zenodo to store the YASA tutorial data. I remembered that Pooch has a nice DOIDownloader class that integrates well with Zenodo to automatically pull metadata about all the files. It's a simpler implementation this way.
Plus I just noticed that YASA also gets archived on Zenodo -- sweet :)
This is good to know and I might squeeze this onto the documentation somewhere. If people want to cite specific versions of YASA that they use in a paper (which is a good idea considering it's early stages of development), I think this is the best way to do it.
@raphaelvallat, I'll create a new YASA tutorial data repository on Zenodo and add you as admin or whatever the highest privileges are.
Accessing/downloading datasets for YASA tutorials would be much easier with Pooch.
OSF orZenodo repository and remove from GitHub.datasets.py
module to access downloadersPooch has very simple and clear documentation for this, and a great resource to work off of is Ensaio.
I'm afraid that some version of this will need to be implemented in
v0.7
. It feels necessary for me to make the tutorial notebooks for the newevaluation
module, since those will require larger datasets to show how their purpose. (therefore kind of blocking #166)**In the future it would be cool to access larger datasets here, like for full analyses rather than demos. It doesn't require any storage. A major benefit would be if YASA performed some standardization step, handling the various formats that people share their PSG data in. But given all the complexities that come with that, this feature should be reserved for a dedicated contributor later on (if interested in implementation at all).
The text was updated successfully, but these errors were encountered: