Programa SAS - IRIS - Machine Learning Supervisionado - An. Discriminante Linear e Quadrático

Programa SAS - IRIS - Machine Learning Supervisionado - An. Discriminante Linear e Quadrático

Programa SAS

Benchmarking Funcional

Programa para Diagnosticar Falsificação de Dinheiro (Franco Suíço):

Analise Discriminante Linear - Funções Canônicas de Fisher

Seminário Gabriel Sarriés e Colegas em Cirso de Machine Learning - Prof. Rafael (Irlanda) .... e Gabriel Machado (Agosto de 2019)

data banco;

input obs Status $ Length Left Right Bottom Top Diagonal;

cards;

1 0 214.8 131.0 131.1 9.0 9.7 141.0

2 0 214.6 129.7 129.7 8.1 9.5 141.7

...

proc discrim data=banco method=normal pool=no

       crossvalidate;

class Status;

priors prop;

var Length Left Right Bottom Top Diagonal;

run;

Fazendo o Benchmarking para IRIS:

data iris;

input Cum_Sep Lar_Sep Cum_Pet Lar_Pet Especie $;

cards;

5.1 3.5 1.4 0.2 I_setosa

4.9 3 1.4 0.2 I_setosa

...

proc discrim data=iris method=normal pool=yes

       crossvalidate;

class Especie;

priors prop;

var Cum_Sep Lar_Sep Cum_Pet Lar_Pet ;

run;

OK Programa Pronto!!!

Observação: Para fazer An. Discriminante Quadrático trocar pool=yes por pool=no, somente temos que fazer isso.

Agora Vamos colocar os dados junto com o Programa que Desenvolvemos por Benchmarking:

Programa SAS para Analise Discriminante Linear para IRIS-Fisher:

data iris;

input Cum_Sep Lar_Sep Cum_Pet Lar_Pet Especie $;

cards;

5.1 3.5 1.4 0.2 I_setosa

4.9 3 1.4 0.2 I_setosa

4.7 3.2 1.3 0.2 I_setosa

4.6 3.1 1.5 0.2 I_setosa

5 3.6 1.4 0.2 I_setosa

5.4 3.9 1.7 0.4 I_setosa

4.6 3.4 1.4 0.3 I_setosa

5 3.4 1.5 0.2 I_setosa

4.4 2.9 1.4 0.2 I_setosa

4.9 3.1 1.5 0.1 I_setosa

5.4 3.7 1.5 0.2 I_setosa

4.8 3.4 1.6 0.2 I_setosa

4.8 3 1.4 0.1 I_setosa

4.3 3 1.1 0.1 I_setosa

5.8 4 1.2 0.2 I_setosa

5.7 4.4 1.5 0.4 I_setosa

5.4 3.9 1.3 0.4 I_setosa

5.1 3.5 1.4 0.3 I_setosa

5.7 3.8 1.7 0.3 I_setosa

5.1 3.8 1.5 0.3 I_setosa

5.4 3.4 1.7 0.2 I_setosa

5.1 3.7 1.5 0.4 I_setosa

4.6 3.6 1 0.2 I_setosa

5.1 3.3 1.7 0.5 I_setosa

4.8 3.4 1.9 0.2 I_setosa

5 3 1.6 0.2 I_setosa

5 3.4 1.6 0.4 I_setosa

5.2 3.5 1.5 0.2 I_setosa

5.2 3.4 1.4 0.2 I_setosa

4.7 3.2 1.6 0.2 I_setosa

4.8 3.1 1.6 0.2 I_setosa

5.4 3.4 1.5 0.4 I_setosa

5.2 4.1 1.5 0.1 I_setosa

5.5 4.2 1.4 0.2 I_setosa

4.9 3.1 1.5 0.1 I_setosa

5 3.2 1.2 0.2 I_setosa

5.5 3.5 1.3 0.2 I_setosa

4.9 3.1 1.5 0.1 I_setosa

4.4 3 1.3 0.2 I_setosa

5.1 3.4 1.5 0.2 I_setosa

5 3.5 1.3 0.3 I_setosa

4.5 2.3 1.3 0.3 I_setosa

4.4 3.2 1.3 0.2 I_setosa

5 3.5 1.6 0.6 I_setosa

5.1 3.8 1.9 0.4 I_setosa

4.8 3 1.4 0.3 I_setosa

5.1 3.8 1.6 0.2 I_setosa

4.6 3.2 1.4 0.2 I_setosa

5.3 3.7 1.5 0.2 I_setosa

5 3.3 1.4 0.2 I_setosa

7 3.2 4.7 1.4 I_versic

6.4 3.2 4.5 1.5 I_versic

6.9 3.1 4.9 1.5 I_versic

5.5 2.3 4 1.3 I_versic

6.5 2.8 4.6 1.5 I_versic

5.7 2.8 4.5 1.3 I_versic

6.3 3.3 4.7 1.6 I_versic

4.9 2.4 3.3 1 I_versic

6.6 2.9 4.6 1.3 I_versic

5.2 2.7 3.9 1.4 I_versic

5 2 3.5 1 I_versic

5.9 3 4.2 1.5 I_versic

6 2.2 4 1 I_versic

6.1 2.9 4.7 1.4 I_versic

5.6 2.9 3.6 1.3 I_versic

6.7 3.1 4.4 1.4 I_versic

5.6 3 4.5 1.5 I_versic

5.8 2.7 4.1 1 I_versic

6.2 2.2 4.5 1.5 I_versic

5.6 2.5 3.9 1.1 I_versic

5.9 3.2 4.8 1.8 I_versic

6.1 2.8 4 1.3 I_versic

6.3 2.5 4.9 1.5 I_versic

6.1 2.8 4.7 1.2 I_versic

6.4 2.9 4.3 1.3 I_versic

6.6 3 4.4 1.4 I_versic

6.8 2.8 4.8 1.4 I_versic

6.7 3 5 1.7 I_versic

6 2.9 4.5 1.5 I_versic

5.7 2.6 3.5 1 I_versic

5.5 2.4 3.8 1.1 I_versic

5.5 2.4 3.7 1 I_versic

5.8 2.7 3.9 1.2 I_versic

6 2.7 5.1 1.6 I_versic

5.4 3 4.5 1.5 I_versic

6 3.4 4.5 1.6 I_versic

6.7 3.1 4.7 1.5 I_versic

6.3 2.3 4.4 1.3 I_versic

5.6 3 4.1 1.3 I_versic

5.5 2.5 4 1.3 I_versic

5.5 2.6 4.4 1.2 I_versic

6.1 3 4.6 1.4 I_versic

5.8 2.6 4 1.2 I_versic

5 2.3 3.3 1 I_versic

5.6 2.7 4.2 1.3 I_versic

5.7 3 4.2 1.2 I_versic

5.7 2.9 4.2 1.3 I_versic

6.2 2.9 4.3 1.3 I_versic

5.1 2.5 3 1.1 I_versic

5.7 2.8 4.1 1.3 I_versic

6.3 3.3 6 2.5 I_virgin

5.8 2.7 5.1 1.9 I_virgin

7.1 3 5.9 2.1 I_virgin

6.3 2.9 5.6 1.8 I_virgin

6.5 3 5.8 2.2 I_virgin

7.6 3 6.6 2.1 I_virgin

4.9 2.5 4.5 1.7 I_virgin

7.3 2.9 6.3 1.8 I_virgin

6.7 2.5 5.8 1.8 I_virgin

7.2 3.6 6.1 2.5 I_virgin

6.5 3.2 5.1 2 I_virgin

6.4 2.7 5.3 1.9 I_virgin

6.8 3 5.5 2.1 I_virgin

5.7 2.5 5 2 I_virgin

5.8 2.8 5.1 2.4 I_virgin

6.4 3.2 5.3 2.3 I_virgin

6.5 3 5.5 1.8 I_virgin

7.7 3.8 6.7 2.2 I_virgin

7.7 2.6 6.9 2.3 I_virgin

6 2.2 5 1.5 I_virgin

6.9 3.2 5.7 2.3 I_virgin

5.6 2.8 4.9 2 I_virgin

7.7 2.8 6.7 2 I_virgin

6.3 2.7 4.9 1.8 I_virgin

6.7 3.3 5.7 2.1 I_virgin

7.2 3.2 6 1.8 I_virgin

6.2 2.8 4.8 1.8 I_virgin

6.1 3 4.9 1.8 I_virgin

6.4 2.8 5.6 2.1 I_virgin

7.2 3 5.8 1.6 I_virgin

7.4 2.8 6.1 1.9 I_virgin

7.9 3.8 6.4 2 I_virgin

6.4 2.8 5.6 2.2 I_virgin

6.3 2.8 5.1 1.5 I_virgin

6.1 2.6 5.6 1.4 I_virgin

7.7 3 6.1 2.3 I_virgin

6.3 3.4 5.6 2.4 I_virgin

6.4 3.1 5.5 1.8 I_virgin

6 3 4.8 1.8 I_virgin

6.9 3.1 5.4 2.1 I_virgin

6.7 3.1 5.6 2.4 I_virgin

6.9 3.1 5.1 2.3 I_virgin

5.8 2.7 5.1 1.9 I_virgin

6.8 3.2 5.9 2.3 I_virgin

6.7 3.3 5.7 2.5 I_virgin

6.7 3 5.2 2.3 I_virgin

6.3 2.5 5 1.9 I_virgin

6.5 3 5.2 2 I_virgin

6.2 3.4 5.4 2.3 I_virgin

5.9 3 5.1 1.8 I_virgin

;

proc discrim data=iris method=normal pool=yes

       crossvalidate;

class Especie;

priors prop;

var Cum_Sep Lar_Sep Cum_Pet Lar_Pet ;

run;

Saída do Programa:
Saída do Programa - Matriz de Cofunção: