[MRG] Minor doc fixes to coordinate_descent.py #11737
Conversation
|
I think the numpydoc spec says something similar |
|
Oh OK, I wasn't aware of that. May I still suggest the following changes?
|
|
Yes, thanks
|
|
this has been discussed in the past. Using non-latex in docstring (not narrative doc) I would not change this. my 2c |
This reverts commit ea05a61.
Removed pseudo latex notation and fixed Fro norm indication
|
I just committed the above proposed changes and changes this PR's title. |
| @@ -1612,13 +1612,13 @@ class MultiTaskElasticNet(Lasso): | |||
|
|
|||
| The optimization objective for MultiTaskElasticNet is:: | |||
|
|
|||
| (1 / (2 * n_samples)) * ||Y - XW||^Fro_2 | |||
| (1 / (2 * n_samples)) * ||Y - XW||_Fro^2 | |||
There was a problem hiding this comment.
+1 for this one, I think it's an error
| + alpha * l1_ratio * ||W||_21 | ||
| + 0.5 * alpha * (1 - l1_ratio) * ||W||_Fro^2 | ||
|
|
||
| Where:: | ||
|
|
||
| ||W||_21 = \\sum_i \\sqrt{\\sum_j w_{ij}^2} | ||
| ||W||_21 = sum_i sqrt(sum_j w_ij ^ 2) |
There was a problem hiding this comment.
Maybe -1 for this one. It seems not friendly to latex users. More importantly, it's not consistent with other parts of the doc (e.g., MultiTaskLasso).
There was a problem hiding this comment.
The comments above suggest that docstrings should not contain any latex.
There was a problem hiding this comment.
I won't be too strict on these minor issues.
Reference Issues/PRs
What does this implement/fix? Explain your changes.
Minor doc changes to use latex instead of plain text math equations.Minor doc changes to remove latex in docstrings + small fix for Frobenius norm index
Any other comments?