[MRG] Modify svm/plot_separating_hyperplane.py example for matplotlib v2 #8369
Conversation
|
I updated your description, this should not close #8364 which is a generic issue about all the examples. Your PR is about one of the examples. |
|
I would advise you to disable CircleCI on your fork unless you have a very good reason not to. There are a variety of reasons for this that I am not going to go into. One is that we have convenient ways to find the generated documentation that do not work if CircleCI is enabled on your fork. |
|
For future reference here are the example links for the stable doc, dev doc and this PR doc. |
| @@ -12,12 +12,13 @@ | |||
| import numpy as np | |||
| import matplotlib.pyplot as plt | |||
| from sklearn import svm | |||
|
|
|||
| from sklearn.datasets import make_classification | |||
There was a problem hiding this comment.
You should probably not change the data unless you have a very good reason to.
|
I modified the title of this PR and description because I am also go through some more examples which needed to change due to matplotlib up-gradation. Matplotlib v2.0 by default support |
Small PRs are way easier to review and merge. Unless the change is a search-and-replace kind of thing (which I don't think it is and on top of it each example will require visual inspection to make sure that everything works fine), I would be in favour of doing it one example at a time. |
|
@lesteve yeah you are right, I do like that . Thanks |
| X = np.r_[np.random.randn(20, 2) - [2, 2], np.random.randn(20, 2) + [2, 2]] | ||
| Y = [0] * 20 + [1] * 20 | ||
|
|
||
| #X = np.r_[np.random.randn(20, 2) - [2, 2], np.random.randn(20, 2) + [2, 2]] |
Codecov Report
@@ Coverage Diff @@
## master #8369 +/- ##
==========================================
+ Coverage 94.75% 95.48% +0.72%
==========================================
Files 342 342
Lines 60809 60913 +104
==========================================
+ Hits 57617 58160 +543
+ Misses 3192 2753 -439
Continue to review full report at Codecov.
|
Plot using contour levels on decision functions
| ylim = ax.get_ylim() | ||
|
|
||
| # create grid to evaluate model | ||
| x = np.linspace(xlim[0], xlim[1], 30) |
There was a problem hiding this comment.
We should really have a function for this but that's another issue #6338
There was a problem hiding this comment.
That's not merge yet.
|
|
||
| X, Y = make_classification(n_features=2, n_redundant=0, n_informative=1, |
There was a problem hiding this comment.
You should use y here. The scikit-learn convention is that variable names starting with a capital letter should be reserved for 2d arrays.
There was a problem hiding this comment.
Can you use a random_state argument otherwise the plot will change each time you run the example.
| plt.scatter(clf.support_vectors_[:, 0], clf.support_vectors_[:, 1], | ||
| s=80, facecolors='none') | ||
| plt.scatter(X[:, 0], X[:, 1], c=Y, cmap=plt.cm.Paired) | ||
| plt.scatter(X[:, 0], X[:, 1], c=Y, s=50, cmap='autumn', edgecolors='k') |
There was a problem hiding this comment.
Do not change something unless there is a very good reason to. In this case, I would keep the plt.cm.Paired colormap.
Reference Issue
#8364
What does this implement/fix? Explain your changes.
Modify scikit-learn examples for better looking plots in matplotlib v2
Any other comments?