Published January 29, 2020
| Version v0.17.1
Software
Open
rasbt/mlxtend: Version 0.17.1
Creators
- Sebastian Raschka1
- James Bourbeau2
- Reiichiro Nakano3
- Zach Griffith4
- Kota Mori
- Will McGinnis5
- JJLWHarrison
- Jakub Šmíd6
- Qiang Gu
- Guillaume Poirier-Morency7
- Daniel
- Floris Hoogenboom
- Colin
- Vahid Mirjalili8
- Steve Harenberg
- Denis Barbier
- Selay
- Christos Aridas
- Pablo Fernandez9
- Oliver Tomic10
- Laurens Geffert
- Janpreet Singh
- Gabriel Azevedo Ferreira
- Alejandro Correa Bahnsen11
- Anton Loss
- Anebi12
- Ajinkya Kale
- Adam Erickson
- Adam Cooper13
- Ackerley Tng14
- 1. UW-Madison
- 2. @Quansight
- 3. @openai
- 4. @WIPACrepo
- 5. Predikto Inc.
- 6. Blindspot Solutions
- 7. @PavlidisLab at Michael Smith Laboratories
- 8. Michigan State University
- 9. FANSI Motorsport
- 10. Norwegian University of Life Sciences
- 11. Easy Solutions
- 12. Eikon Labs
- 13. Tribal
- 14. Centre for Strategic Infocomm Technologies
Description
New Features
- The
SequentialFeatureSelector
now supports using pre-specified feature sets via thefixed_features
parameter. (#578) - Adds a new
accuracy_score
function tomlxtend.evaluate
for computing basic classifcation accuracy, per-class accuracy, and average per-class accuracy. (#624 via Deepan Das) StackingClassifier
andStackingCVClassifier
now have adecision_function
method, which serves as a preferred choice overpredict_proba
in calculating roc_auc and average_precision scores when the meta estimator is a linear model or support vector classifier. (#634 via Qiang Gu)
- Improve the runtime performance for the
apriori
frequent itemset generating function whenlow_memory=True
. Settinglow_memory=False
(default) is still faster for small itemsets, butlow_memory=True
can be much faster for large itemsets and requires less memory. Also, input validation forapriori
, ̀ fpgrowthand
fpmaxtakes a significant amount of time when input pandas DataFrame is large; this is now dramatically reduced when input contains boolean values (and not zeros/ones), which is the case when using
TransactionEncoder`. (#619 via Denis Barbier) - Add support for newer sparse pandas DataFrame for frequent itemset algorithms. Also, input validation for
apriori
, ̀ fpgrowthand
fpmax` runs much faster on sparse DataFrame when input pandas DataFrame contains integer values. (#621 via Denis Barbier) - Let
fpgrowth
andfpmax
directly work on sparse DataFrame, they were previously converted into dense Numpy arrays. (#622 via Denis Barbier)
- Fixes a bug in
mlxtend.plotting.plot_pca_correlation_graph
that caused the explaind variances not summing up to 1. Also, improves the runtime performance of the correlation computation and adds a missing function argument for the explained variances (eigenvalues) if users provide their own principal components. (#593 via Gabriel Azevedo Ferreira) - Behavior of
fpgrowth
andapriori
consistent for edgecases such asmin_support=0
. (#573 via Steve Harenberg) fpmax
returns an empty data frame now instead of raising an error if the frequent itemset set is empty. (#573 via Steve Harenberg)- Fixes and issue in
mlxtend.plotting.plot_confusion_matrix
, where the font-color choice for medium-dark cells was not ideal and hard to read. #588 via sohrabtowfighi) - The
svd
mode ofmlxtend.feature_extraction.PrincipalComponentAnalysis
now also n-1 degrees of freedom instead of n d.o.f. when computing the eigenvalues to match the behavior ofeigen
. #595 - Disable input validation for
StackingCVClassifier
because it causes issues if pipelines are used as input. #606
Files
rasbt/mlxtend-v0.17.1.zip
Files
(16.3 MB)
Name | Size | Download all |
---|---|---|
md5:57b9bb87a16585f05247bdd71d517b50
|
16.3 MB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/rasbt/mlxtend/tree/v0.17.1 (URL)