Example 2D Visualizations Using the Iris Data



In [1]:

    
%pylab inline
from classy import *









    



Populating the interactive namespace from numpy and matplotlib
Keras not installed
Version:  0.0.19



In [2]:

    
data=load_excel('data/iris.xls',verbose=True)









    



iris.data 151 5
150 vectors of length 4
Feature names: 'petal length in cm', 'petal width in cm', 'sepal length in cm', 'sepal width in cm'
Target values given.
Target names: 'Iris-setosa', 'Iris-versicolor', 'Iris-virginica'
Mean:  [3.75866667 1.19866667 5.84333333 3.054     ]
Median:  [4.35 1.3  5.8  3.  ]
Stddev:  [1.75852918 0.76061262 0.82530129 0.43214658]

All possible 2D plots for features



In [3]:

    
plot_feature_combinations(data)

or choose which features to plot...



In [4]:

    
plot_feature_combinations(data,['petal length in cm','sepal width in cm','petal width in cm'])

Naive Bayes

since we're only interesting in viewing here, let's train with all the data, but just two features.



In [21]:

    
data_train=extract_features(data,[2,3])



In [22]:

    
C=NaiveBayes()



In [23]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print("Training time: ",timeit())









    



Time Reset
Training time:  0.0025658607482910156 seconds



In [24]:

    
figure(figsize=(12,8))
plot2D(data_train,C,axis_range=[0,10.5,0,8])
C.plot_centers()  # plot the classifier information

kNearestNeighbor



In [25]:

    
C=kNearestNeighbor()



In [26]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print("Training time: ",timeit())









    



Time Reset
Training time:  0.0015039443969726562 seconds



In [27]:

    
figure(figsize=(12,8))
plot2D(data_train,C,axis_range=[0,10.5,0,8])

RCE



In [28]:

    
C=RCE()



In [29]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print("Training time: ",timeit())









    



Time Reset
Training time:  0.27827882766723633 seconds



In [30]:

    
print("number of centers:",len(C.centers))









    



number of centers: 76



In [31]:

    
figure(figsize=(12,8))
C.plot_centers()
plot2D(data_train,C,axis_range=[0,10.5,-1,8])

CSC



In [32]:

    
C=CSC()



In [33]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print("Training time: ",timeit())









    



Time Reset
Training time:  0.03620004653930664 seconds



In [34]:

    
figure(figsize=(12,8))
C.plot_centers()
plot2D(data_train,C,axis_range=[0,10.5,-1,8])

Perceptron



In [ ]:

    
C=Perceptron()



In [35]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print(("Training time: ",timeit()))









    



Time Reset
('Training time: ', '0.03583693504333496 seconds ')



In [36]:

    
figure(figsize=(12,8))
plot2D(data_train,C,axis_range=[0,10.5,-1,8])

Backprop



In [44]:

    
C=BackProp(hidden_layer_sizes = [4],max_iter=10000,tol=1e-4)



In [45]:

    
timeit(reset=True)
C.fit(data_train.vectors,data_train.targets)
print(("Training time: ",timeit()))









    



Time Reset
('Training time: ', '0.7550270557403564 seconds ')



In [46]:

    
figure(figsize=(12,8))
plot2D(data_train,C,axis_range=[0,10.5,-1,8])



In [ ]: