discrimintools.summaryMDA#

discrimintools.summaryMDA(obj, digits=4, detailed=False, to_markdown=False, tablefmt='github', **kwargs)[source]#

Printing summaries of Mixed Discriminant Analysis model.

Parameters:

  • obj (class) – An object of class MDA.

  • digits (int, default = 4) – The number of decimal printed.

  • detailed (bool, default = False) – To print detailed summaries.

  • to_markdown (bool, default = False) – To print summaries in markdown-friendly format. Requires the tabulate. package.

  • tablefmt (str, default = “github”) – The table format.

  • **kwargs – additionals parameters. These parameters will be passed to tabulate.

Return type:

NoneType

See also

summaryCANDISC

Printing summaries of Canonical Discriminant Analysis model.

summaryCPLS

Printing summaries of Partial Least Squares for Classification model.

summaryDA

Printing summaries of Discriminant Analysis model.

summaryDiCA

Printing summaries of Discriminant Correspondence Analysis model.

summaryDISCRIM

Printing summaries of Discriminant Analysis (linear and quadratic) model.

summaryGFALDA

Printing summaries of General Factor Analysis Linear Discriminant Analysis model.

summaryPLSDA

Printing summaries of Partial Least Squares Discriminant Analysis model.

summaryPLSLDA

Printing summaries of Partial Least Squares Linear Discriminant Analysis model.

summarySTEPDISC

Printing summaries of Stepwise Discriminant Analysis model.

Examples

>>> from discrimintools.datasets import load_heart
>>> from discrimintools import MDA, summaryMDA
>>> D = load_heart("subset") # load training data
>>> y, X = D["disease"], D.drop(columns=["disease"]) # split into X and y
>>> clf = MDA(n_components=5)
>>> clf.fit(X,y)
MDA(n_components=5)
>>> summaryMDA(clf)
                    Mixed Discriminant Analysis - Results
Importance of components:
      Eigenvalue  Difference  Proportion (%)  Cumulative (%)
Can1      3.5040      1.4985         25.0288         25.0288
Can2      2.0055      0.1059         14.3249         39.3536
Can3      1.8996      0.5750         13.5684         52.9220
Can4      1.3246      0.1522          9.4616         62.3837
Can5      1.1724      0.0681          8.3743         70.7580
Raw Canonical Coefficients:
                         Can1    Can2    Can3    Can4    Can5
Constant              -0.0059  0.2104  0.2151 -0.8402 -5.4678
age                   -0.0333  0.0137  0.0030  0.0189  0.0549
restbpress            -0.0109  0.0047  0.0056 -0.0012  0.0297
max_hrate              0.0150  0.0020  0.0027  0.0007 -0.0121
asympt                -0.7662 -0.3278 -0.4341 -0.0563 -0.5999
...
left_vent_hyper        0.1982  0.5016  1.7704  1.4714 -1.7099
normal                 0.4200  0.4352 -1.7807 -0.1179  0.2904
st_t_wave_abnormality -0.5140 -0.5888  1.6832 -0.1457 -0.0045
no                     0.9448  0.1881  0.2183  0.1125  0.1830
yes                   -0.9448 -0.1881 -0.2183 -0.1125 -0.1830
Projection functions coefficients:
                         Can1    Can2    Can3    Can4    Can5
age                   -0.0892  0.0366  0.0081  0.0507  0.1472
restbpress            -0.0634  0.0270  0.0322 -0.0067  0.1723
max_hrate              0.1190  0.0162  0.0218  0.0054 -0.0965
asympt                -0.0957 -0.0410 -0.0542 -0.0070 -0.0750
atyp_angina            0.0676  0.0221  0.0450 -0.1487  0.0679
...
left_vent_hyper        0.0076  0.0192  0.0676  0.0562 -0.0653
normal                 0.0392  0.0406 -0.1662 -0.0110  0.0271
st_t_wave_abnormality -0.0451 -0.0516  0.1475 -0.0128 -0.0004
no                     0.1122  0.0224  0.0259  0.0134  0.0217
yes                   -0.1122 -0.0224 -0.0259 -0.0134 -0.0217
On this page

This Page