[Enh] add example and handle one-dimensional arrays in CSD.generate lfp #594
Conversation
Moritz-Alexander-Kern
added
documentation
Co-authored-by: Alexander Kleinjohann <[email protected]>
Co-authored-by: Alexander Kleinjohann <[email protected]>
| @@ -253,10 +253,31 @@ def generate_lfp(csd_profile, x_positions, y_positions=None, z_positions=None, | |||
|
|
|||
| Returns | |||
| ------- | |||
| LFP : neo.AnalogSignal | |||
| LFP : :class:`neo.core.AnalogSignal` | |||
There was a problem hiding this comment.
This change should also apply in line 74, right?
(Can't directly suggest the change here, because it's too far from any lines changed in this PR.
There was a problem hiding this comment.
Indeed, thanks for catching this, changed accordingly.
elephant/current_source_density.py
Outdated
| @@ -227,7 +227,7 @@ def generate_lfp(csd_profile, x_positions, y_positions=None, z_positions=None, | |||
| 1D : gauss_1d_dipole | |||
| 2D : large_source_2D and small_source_2D | |||
| 3D : gauss_3d_dipole | |||
| x_positions : np.ndarray | |||
| x_positions : np.ndarray # TODO: add dimensionality for x-y-z_positions | |||
There was a problem hiding this comment.
I assume this TODO should be addressed in another future PR and not here?
There was a problem hiding this comment.
Thank you for the reminder, I should have made this TODO more explicit.
One might consider to further discuss the expected input shape in a future PR or Issue (?)
I've now (fe7f9ec) updated the docstring to explicitly specify the expected shape of the inputs as a 2D column vector with dimensions 'N x 1 array.' In the example mentioned in the issue, an error occurred because a flat array (1D) was passed as input.
For now, documenting the expected input is a reasonable solution ?
Moritz-Alexander-Kern
merged commit 913a998
into
NeuralEnsemble:master
This PR closes #546 .
Changes
x_positions