agastidukare

result = classifier(zipped_clf, x_train, y_train, x_test, y_test)

agastidukare

def fit_classifier(pipeline, x_train, y_train, x_test, y_test):
    model_fit = pipeline.fit(x_train, y_train)
    y_pred = model_fit.predict(x_test)
    accuracy = accuracy_score(y_test, y_pred)
    print("accuracy score: {0:.2f}%".format(accuracy*100))
    return accuracy

agastidukare

def classifier(classifier, t_train, c_train, t_test, c_test):
    result = []
    for n,c in classifier:
        checker_pipeline = Pipeline([
            ('standardize', StandardScaler()),                         
            ('classifier', c)
        ])
        print("Validation result for {}".format(n))
        print(c)
        clf_acc = fit_classifier(checker_pipeline, t_train, c_train, t_test,c_test)

agastidukare

classifier_names = ["Logistic Regression", "KNN", "Random Forest","SVM"]

classifiers = [LogisticRegression(), KNeighborsClassifier(), RandomForestClassifier(), LinearSVC()]

zipped_clf = zip(classifier_names,classifiers)

agastidukare

iris = datasets.load_iris()
x = iris.data
y = iris.target
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.33, random_state=42)

agastidukare

from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import Pipeline
from sklearn.metrics import accuracy_score

agastidukare

# Load perturbted image and print predictions after pertubation 
test_images = per_img
test_labels = label 
grads,acc = covnet(1,params)

agastidukare

# This function calculates perturbted image
def perturb(grad,img,ep):
  grads = grad[0]
  a = numpy.reshape(grads, (1, 784))
  s = np.sign(a)
  perturbed_image = img + np.dot(ep,s)
  return perturbed_image
per_img = perturb(grads,img,0.3)
display(per_img)

agastidukare

# load desired image and its label in test set
def load_img(image,img_label):
  img = np.array(image)
  img = img.reshape(1,784)
  label = np.array(img_label)
  label = label.reshape(1,10)
  return img, label

img, label = load_img(test_images[0],test_labels[0])  
test_images = img

agastidukare

def display(image):
  img = image[0].reshape((28,28))
  plt.imshow(img, cmap="Greys")
  plt.show()
  return

display(a)