supereeg

supereeg copied to clipboard

We need to pin specific versions for pandas >= 1.1.0,

lucywowen

 Slicing a brain object with an int variable produces an empty brain object (no elecs, no data). Slicing a brain object with...

pmdaly

Issue with downsampling nifti images. This fixed it: `from nilearn.image import resample_img` `downsampled_nii = resample_img(yeo, target_affine=np.eye(3)*4., interpolation='nearest')`

lucywowen

https://streamlit.io/ for the client-side aws serverless architechture for the backend - leveraging scaling lambda, and dynamodb, built on flask for easy api development and interoperability with supereeg https://serverless.com/blog/flask-python-rest-api-serverless-lambda-dynamodb/ how to...

tmuntianu

Add a new `event` feature. This would be implemented similar to `session` variable and would serve as a tag/label for tagging event onsets, behavioral and experimental events . Also add...

lucywowen

- for each channel calculate kurtosis using a sliding window - plot kurtosis values - try this in ipyvolume

lucywowen

lucywowen

for next release, change nifti get_slice method to something like this: `def get_slice(self, index):` ` if len(self.shape)>3:` ` if self.shape[3]==1:` ` return nib.Nifti1Image(header=self.header, dataobj=np.squeeze(self.get_data()), affine=self.affine)`

lucywowen

API ideas brainstorm

2

**Nifti:** - Add a Nifti.get_locs() function: - One idea: first convert to a brain object, and then return bo.get_locs(). - Another idea: compute these locations when we initialize the model,...

jeremymanning

This division when the model is recovered should probably be done in log space. Think further on how to do this.

lucywowen