[MRG] ENH Cache enabling in Pipeline #281
Conversation
Codecov Report
@@ Coverage Diff @@
## master #281 +/- ##
==========================================
+ Coverage 98.27% 98.32% +0.04%
==========================================
Files 58 58
Lines 3427 3703 +276
==========================================
+ Hits 3368 3641 +273
- Misses 59 62 +3
Continue to review full report at Codecov.
|
|
@chkoar you can already have a look. I have to check if there is any issue with the caching of the sampler. |
glemaitre
changed the title
|
@chkoar I added the missing test |
imblearn/pipeline.py
Outdated
| from sklearn.externals import six | ||
| from sklearn.externals.joblib import Memory | ||
| from sklearn.utils import tosequence | ||
| from sklearn.utils.metaestimators import if_delegate_has_method | ||
|
|
||
| __all__ = ['Pipeline'] |
There was a problem hiding this comment.
Could you add make_pipeline because I have forgotten?
|
That is a great addition Guillaume! I don't have the time to review this great PR but I think that since the tests are passing we could easily merge it. |
|
I took the scikit-learn test and add your additional test as well. so it should be fine. |
* ENH add cache for transformer with test from scikit-learn * DOC improve the doc * TST add sampler test * ENH add make_pipeline in the import
* ENH add cache for transformer with test from scikit-learn * DOC improve the doc * TST add sampler test * ENH add make_pipeline in the import
Reference Issue
Fixed #280
What does this implement/fix? Explain your changes.
Allow for caching Transformer and Sampler within a Pipeline.
Any other comments?