clasifier python

[yolobot]

πως μπορω να καλεσω εναν classifier perception και να τα απεικονισω σε ενα sublot 2x2?

[Dr.Fuzzy]

βιβλιοθήκη; υλοποίηση; Η ερώτηση σου είναι πάρα πολύ γενική.

[yolobot]

εχω τον παρακατω κωδικα και θελω να κανω sublot  ωστε να κανω  4  διαφορετικά πειράματα με

διαφορετική ποσόστωση μεταξύ train και test δεδομένων (10%, 20%,
30%, 40%) και  να απεικονίσω τις αποφάσεις που πήρε ο classifier σε ένα
subplot 2x2

# -*- coding: utf-8 -*-
"""

"""
from sklearn import datasets
import numpy as np

iris = datasets.load_iris() # fortosi dataset
print 'ena mikro deigma tou dataset'
print 'attribute values = ' , iris.data[0:5]
print 'class values = ' , iris.target[0:5]
print 'onomasia twn atributes'
print iris.feature_names
print 'onomasia twn katigoriwn'
print iris.target_names
print ' '
print 'diastaseis tou dataset = ' , iris.data.shape
print type(iris), type(iris.data), type(iris.target)

# xrhsh mono ton xaraktiristikon 3 kai 4 (2,3 gia python)
X = iris.data[:, [2, 3]]
y = iris.target
print ' '
print('Class labels:', np.unique(y))

# xorismos dedomenon se 70% train kai 30% test
from sklearn.cross_validation import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
         X, y, test_size = 0.1, random_state = 0)

X_train, X_test, y_train, y_test = train_test_split(
         X, y, test_size = 0.2, random_state = 0)        
        
       
# kanonikopoihsh ton dedomenon
from sklearn.preprocessing import StandardScaler

sc = StandardScaler()
sc.fit(X_train)
X_train_std = sc.transform(X_train)
X_test_std = sc.transform(X_test)

from sklearn.linear_model import Perceptron

ppn = Perceptron(n_iter=100, eta0=0.1, random_state=0)
ppn.fit(X_train_std, y_train)
y_test.shape
y_pred = ppn.predict(X_test_std)
print('Misclassified samples: %d' % (y_test != y_pred).sum())

from sklearn.metrics import accuracy_score

print('Accuracy: %.2f' % accuracy_score(y_test, y_pred))

from matplotlib.colors import ListedColormap
import matplotlib.pyplot as plt

#matplotlib inline

def plot_decision_regions(X, y, classifier, test_idx=None, resolution=0.02):

    # setup marker generator and color map
    markers = ('s', 'x', 'o', '^', 'v')
    colors = ('red', 'blue', 'lightgreen', 'gray', 'cyan')
    cmap = ListedColormap(colors[:len(np.unique(y))])

    # plot the decision surface
    x1_min, x1_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    x2_min, x2_max = X[:, 1].min() - 1, X[:, 1].max() + 1
    xx1, xx2 = np.meshgrid(np.arange(x1_min, x1_max, resolution),
                         np.arange(x2_min, x2_max, resolution))
    Z = classifier.predict(np.array([xx1.ravel(), xx2.ravel()]).T)
    Z = Z.reshape(xx1.shape)
    plt.contourf(xx1, xx2, Z, alpha=0.4, cmap=cmap)
    plt.xlim(xx1.min(), xx1.max())
    plt.ylim(xx2.min(), xx2.max())

    # plot all samples
    X_test, y_test = X[test_idx, :], y[test_idx]                              
    for idx, cl in enumerate(np.unique(y)):
        plt.scatter(x=X[y == cl, 0], y=X[y == cl, 1],
                    alpha=0.8, c=cmap(idx),
                    marker=markers[idx], label=cl)
       
    # highlight test samples
    if test_idx:
        X_test, y_test = X[test_idx, :], y[test_idx]  
        plt.scatter(X_test[:, 0], X_test[:, 1], c='',
                alpha=1.0, linewidth=1, marker='o',
                s=55, label='test set')

X_combined_std = np.vstack((X_train_std, X_test_std))
y_combined = np.hstack((y_train, y_test))

plot_decision_regions(X=X_combined_std, y=y_combined,
                      classifier=ppn, test_idx=range(105,150))
#plt.figure(num=None, figsize=(10,8 ), dpi=300, facecolor='w', edgecolor='k')
                    
plt.xlabel('petal length [standardized]' , fontsize = 20)
plt.ylabel('petal width [standardized]', fontsize = 20)
plt.legend(loc='upper left' , fontsize = 20)
plt.title('100 iterations eta = 0.1', fontsize = 24)

 

plt.tight_layout()
# plt.savefig('./figures/iris_perceptron_scikit.png', dpi=300)
plt.show()

[Dr.Fuzzy]

Δες εδώ έχει ότι χρειάζεσαι http://matplotlib.org/examples/pylab_examples/subplots_demo.html

[yolobot]

πως θα βαλω να τα εμφανιζει μεσα στα 4 παραθυρα που μου εμφανιζει ασπρα απλα αλλαζοντας το test size 0.1, 0.2, 0.3, .04

μου εμφανιζει αυτο τωρα ιδου και ο παρακατω κωδικας

# -*- coding: utf-8 -*-

"""
@author: stkarlos
"""
from sklearn import datasets
import numpy as np

iris = datasets.load_iris() # fortosi dataset
print 'ena mikro deigma tou dataset'
print 'attribute values = ' , iris.data[0:5]
print 'class values = ' , iris.target[0:5]
print 'onomasia twn atributes'
print iris.feature_names
print 'onomasia twn katigoriwn'
print iris.target_names
print ' '
print 'diastaseis tou dataset = ' , iris.data.shape
print type(iris), type(iris.data), type(iris.target)

# xrhsh mono ton xaraktiristikon 3 kai 4 (2,3 gia python)
X = iris.data[:, [2, 3]]
y = iris.target
print ' '
print('Class labels:', np.unique(y))

# xorismos dedomenon se 70% train kai 30% test
from sklearn.cross_validation import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
         X, y, test_size = 0.3, random_state = 0)
       
# kanonikopoihsh ton dedomenon
from sklearn.preprocessing import StandardScaler

sc = StandardScaler()
sc.fit(X_train)
X_train_std = sc.transform(X_train)
X_test_std = sc.transform(X_test)

from sklearn.linear_model import Perceptron

ppn = Perceptron(n_iter=100, eta0=0.1, random_state=0)
ppn.fit(X_train_std, y_train)
y_test.shape
y_pred = ppn.predict(X_test_std)
print('Misclassified samples: %d' % (y_test != y_pred).sum())

from sklearn.metrics import accuracy_score

print('Accuracy: %.2f' % accuracy_score(y_test, y_pred))

from matplotlib.colors import ListedColormap
import matplotlib.pyplot as plt

#matplotlib inline

def plot_decision_regions(X, y, classifier, test_idx=None, resolution=0.02):

    # setup marker generator and color map
    markers = ('s', 'x', 'o', '^', 'v')
    colors = ('red', 'blue', 'lightgreen', 'gray', 'cyan')
    cmap = ListedColormap(colors[:len(np.unique(y))])

    # plot the decision surface
    x1_min, x1_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    x2_min, x2_max = X[:, 1].min() - 1, X[:, 1].max() + 1
    xx1, xx2 = np.meshgrid(np.arange(x1_min, x1_max, resolution),
                         np.arange(x2_min, x2_max, resolution))
    Z = classifier.predict(np.array([xx1.ravel(), xx2.ravel()]).T)
    Z = Z.reshape(xx1.shape)
    plt.contourf(xx1, xx2, Z, alpha=0.4, cmap=cmap)
    plt.xlim(xx1.min(), xx1.max())
    plt.ylim(xx2.min(), xx2.max())

    # plot all samples
    X_test, y_test = X[test_idx, :], y[test_idx]                              
    for idx, cl in enumerate(np.unique(y)):
        plt.scatter(x=X[y == cl, 0], y=X[y == cl, 1],
                    alpha=0.8, c=cmap(idx),
                    marker=markers[idx], label=cl)
       
    # highlight test samples
    if test_idx:
        X_test, y_test = X[test_idx, :], y[test_idx]  
        plt.scatter(X_test[:, 0], X_test[:, 1], c='',
                alpha=1.0, linewidth=1, marker='o',
                s=55, label='test set')

X_combined_std = np.vstack((X_train_std, X_test_std))
y_combined = np.hstack((y_train, y_test))

plot_decision_regions(X=X_combined_std, y=y_combined,
                      classifier=ppn, test_idx=range(105,150))
#plt.figure(num=None, figsize=(10,8 ), dpi=300, facecolor='w', edgecolor='k')
# Four axes, returned as a 2-d array
f, axarr = plt.subplots(2, 2)
axarr[0, 0].plot(x, y)
axarr[0, 0].set_title('Axis [0,0]')
axarr[0, 1].scatter(x, y)
axarr[0, 1].set_title('Axis [0,1]')
axarr[1, 0].plot(x, y ** 2)
axarr[1, 0].set_title('Axis [1,0]')
axarr[1, 1].scatter(x, y ** 2)
axarr[1, 1].set_title('Axis [1,1]')
# Fine-tune figure; hide x ticks for top plots and y ticks for right plots
plt.setp([a.get_xticklabels() for a in axarr[0, :]], visible=False)
plt.setp([a.get_yticklabels() for a in axarr[:, 1]], visible=False)
plt.xlabel('petal length [standardized]' , fontsize = 20)
plt.ylabel('petal width [standardized]', fontsize = 20)
plt.legend(loc='upper left' , fontsize = 20)
plt.title('100 iterations eta = 0.1', fontsize = 24)

plt.tight_layout()
# plt.savefig('./figures/iris_perceptron_scikit.png', dpi=300)
plt.show()

[yolobot]

βοηθεια καποιος?