[MRG] DOC: Adding documentation for the argument X in fit() #7784
Conversation
| @@ -875,6 +875,7 @@ def fit(self, X, y=None): | |||
| Parameters | |||
| ---------- | |||
| X : array-like or sparse matrix, shape=(n_samples, n_features) | |||
| Coordinates of training data points to cluster | |||
There was a problem hiding this comment.
We usually refer to X as the features of the observations, not the coordinates (they could be colors, grades, blood sugar level… pretty much anything).
There was a problem hiding this comment.
@NelleV Correctly stated. I made a mistake there. How about extending 'sparse matrix with each element representing the feature vector for the training data points to cluster'?
|
Sorry, this ticket has been addressed in another merged pull request (#7783) |
|
I just merged #7783 as it uses a wording that I feel is more common in scikit-learn, and therefore is less surprising. I hope you don't mind. In general, it's good to coordinate with other people that already opened PRs. Thank you for your work! |
|
Yes Sir, I am just a beginner at open source. Thank you for letting me know. I'll keep that in mind for future issues. |
Reference Issue
Addressing #7772
What does this implement/fix? Explain your changes.
This adds a more clear description for the argument X of the function fit() here.
Any other comments?
Since we are computing the centroid, using 'Coordinates' seems more intuitive. Also, this example here helps to understand how X could be visualized as a point in a d-dimensional space.