Discussion:
[pystatsmodels] Confusion matrix statistical analysis
Sepand Haghighi
2018-12-07 14:25:10 UTC
Permalink
Dear All,

Here I want to introduce an open source Python library which named PyCM.
PyCM is a machine learning library providing statistical analysis of
confusion matrix through a large variety of parameters such as AUC,
Confusion Entropy, information theory related parameters, and etc. This
developing library can be used in order to evaluate the performance of
different machine learning algorithms by offering different evaluation
parameters on their resulted confusion matrix.

PyCM is a multi-class confusion matrix library written in Python that
supports both input data vectors and direct matrix, and a proper tool for
post-classification model evaluation that supports most classes and overall
statistics parameters. PyCM is the swiss-army knife of confusion matrices,
targeted mainly at data scientists that need a broad array of metrics for
predictive models and an accurate evaluation of large variety of
classifiers.

Do not hesitate to contact us about this library and help us to develop it
by your valuable suggestions.
You can find us on https://github.com/sepandhaghighi/pycm
<http://dear%20all%2C%20%20%20here%20i%20want%20to%20introduce%20an%20open%20source%20python%20library%20which%20named%20pycm.%20pycm%20is%20a%20machine%20learning%20library%20providing%20statistical%20analysis%20of%20confusion%20matrix%20through%20a%20large%20variety%20of%20parameters%20such%20as%20auc%2C%20confusion%20entropy%2C%20information%20theory%20related%20parameters%2C%20and%20etc.%20this%20developing%20library%20can%20be%20used%20in%20order%20to%20evaluate%20the%20performance%20of%20different%20machine%20learning%20algorithms%20by%20offering%20different%20evaluation%20parameters%20on%20their%20resulted%20confusion%20matrix.%20%20pycm%20is%20a%20multi-class%20confusion%20matrix%20library%20written%20in%20python%20that%20supports%20both%20input%20data%20vectors%20and%20direct%20matrix%2C%20and%20a%20proper%20tool%20for%20post-classification%20model%20evaluation%20that%20supports%20most%20classes%20and%20overall%20statistics%20parameters.%20pycm%20is%20the%20swiss-army%20knife%20of%20confusion%20matrices%2C%20targeted%20mainly%20at%20data%20scientists%20that%20need%20a%20broad%20array%20of%20metrics%20for%20predictive%20models%20and%20an%20accurate%20evaluation%20of%20large%20variety%20of%20classifiers.%20%20do%20not%20hesitate%20to%20contact%20us%20about%20this%20library%20and%20help%20us%20to%20develop%20it%20by%20your%20valuable%20suggestions.%20you%20can%20find%20us%20on%20https//github.com/sepandhaghighi/pycm>
j***@gmail.com
2018-12-07 15:40:55 UTC
Permalink
Post by Sepand Haghighi
Dear All,
Here I want to introduce an open source Python library which named PyCM.
PyCM is a machine learning library providing statistical analysis of
confusion matrix through a large variety of parameters such as AUC,
Confusion Entropy, information theory related parameters, and etc. This
developing library can be used in order to evaluate the performance of
different machine learning algorithms by offering different evaluation
parameters on their resulted confusion matrix.
Post by Sepand Haghighi
PyCM is a multi-class confusion matrix library written in Python that
supports both input data vectors and direct matrix, and a proper tool for
post-classification model evaluation that supports most classes and overall
statistics parameters. PyCM is the swiss-army knife of confusion matrices,
targeted mainly at data scientists that need a broad array of metrics for
predictive models and an accurate evaluation of large variety of
classifiers.
Post by Sepand Haghighi
Do not hesitate to contact us about this library and help us to develop
it by your valuable suggestions.
Post by Sepand Haghighi
You can find us on
https://github.com/sepandhaghighi/pycm

Thanks for the package and the announcement. It has an impressive list of
statistics.

The related older statistical literature is mostly under the labels
interrater reliability and agreement literature in psychology and several
other fields

Statsmodels has only two or three measures and several open issues for
extensions to it
http://www.statsmodels.org/devel/stats.html#interrater-reliability-and-agreement
https://github.com/statsmodels/statsmodels/issues/4387

As I mentioned in a recent mailing list message, I would like to link to
packages like this directly from the statsmodels TOC, and copy/translate a
few of the functions that are most popular in statistics to statsmodels.

Josef

Continue reading on narkive:
Loading...