Chapter 3 – Classification
This notebook contains all the sample code and solutions to the exercises in chapter 3.
First, let's make sure this notebook works well in both python 2 and 3, import a few common modules, ensure MatplotLib plots figures inline and prepare a function to save the figures:
# To support both python 2 and python 3
from __future__ import division, print_function, unicode_literals
# Common imports
import numpy as np
import os
# to make this notebook's output stable across runs
np.random.seed(42)
# To plot pretty figures
%matplotlib inline
import matplotlib
import matplotlib.pyplot as plt
plt.rcParams['axes.labelsize'] = 14
plt.rcParams['xtick.labelsize'] = 12
plt.rcParams['ytick.labelsize'] = 12
# Where to save the figures
PROJECT_ROOT_DIR = "."
CHAPTER_ID = "classification"
def save_fig(fig_id, tight_layout=True):
path = os.path.join(PROJECT_ROOT_DIR, "images", CHAPTER_ID, fig_id + ".png")
print("Saving figure", fig_id)
if tight_layout:
plt.tight_layout()
plt.savefig(path, format='png', dpi=300)
from sklearn.datasets import fetch_mldata
mnist = fetch_mldata('MNIST original')
mnist
{'COL_NAMES': ['label', 'data'], 'DESCR': 'mldata.org dataset: mnist-original', 'data': array([[0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0], ..., [0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0]], dtype=uint8), 'target': array([ 0., 0., 0., ..., 9., 9., 9.])}
X, y = mnist["data"], mnist["target"]
X.shape
(70000, 784)
y.shape
(70000,)
28*28
784
%matplotlib inline
import matplotlib
import matplotlib.pyplot as plt
some_digit = X[36000]
some_digit_image = some_digit.reshape(28, 28)
plt.imshow(some_digit_image, cmap = matplotlib.cm.binary,
interpolation="nearest")
plt.axis("off")
save_fig("some_digit_plot")
plt.show()
Saving figure some_digit_plot
def plot_digit(data):
image = data.reshape(28, 28)
plt.imshow(image, cmap = matplotlib.cm.binary,
interpolation="nearest")
plt.axis("off")
# EXTRA
def plot_digits(instances, images_per_row=10, **options):
size = 28
images_per_row = min(len(instances), images_per_row)
images = [instance.reshape(size,size) for instance in instances]
n_rows = (len(instances) - 1) // images_per_row + 1
row_images = []
n_empty = n_rows * images_per_row - len(instances)
images.append(np.zeros((size, size * n_empty)))
for row in range(n_rows):
rimages = images[row * images_per_row : (row + 1) * images_per_row]
row_images.append(np.concatenate(rimages, axis=1))
image = np.concatenate(row_images, axis=0)
plt.imshow(image, cmap = matplotlib.cm.binary, **options)
plt.axis("off")
plt.figure(figsize=(9,9))
example_images = np.r_[X[:12000:600], X[13000:30600:600], X[30600:60000:590]]
plot_digits(example_images, images_per_row=10)
save_fig("more_digits_plot")
plt.show()
Saving figure more_digits_plot
y[36000]
5.0
X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]
import numpy as np
shuffle_index = np.random.permutation(60000)
X_train, y_train = X_train[shuffle_index], y_train[shuffle_index]
y_train_5 = (y_train == 5)
y_test_5 = (y_test == 5)
from sklearn.linear_model import SGDClassifier
sgd_clf = SGDClassifier(random_state=42)
sgd_clf.fit(X_train, y_train_5)
/home/ageron/.virtualenvs/ml/lib/python3.5/site-packages/sklearn/linear_model/stochastic_gradient.py:84: FutureWarning: max_iter and tol parameters have been added in <class 'sklearn.linear_model.stochastic_gradient.SGDClassifier'> in 0.19. If both are left unset, they default to max_iter=5 and tol=None. If tol is not None, max_iter defaults to max_iter=1000. From 0.21, default max_iter will be 1000, and default tol will be 1e-3. "and default tol will be 1e-3." % type(self), FutureWarning)
SGDClassifier(alpha=0.0001, average=False, class_weight=None, epsilon=0.1, eta0=0.0, fit_intercept=True, l1_ratio=0.15, learning_rate='optimal', loss='hinge', max_iter=5, n_iter=None, n_jobs=1, penalty='l2', power_t=0.5, random_state=42, shuffle=True, tol=None, verbose=0, warm_start=False)
sgd_clf.predict([some_digit])
array([ True], dtype=bool)
from sklearn.model_selection import cross_val_score
cross_val_score(sgd_clf, X_train, y_train_5, cv=3, scoring="accuracy")
array([ 0.9502 , 0.96565, 0.96495])
from sklearn.model_selection import StratifiedKFold
from sklearn.base import clone
skfolds = StratifiedKFold(n_splits=3, random_state=42)
for train_index, test_index in skfolds.split(X_train, y_train_5):
clone_clf = clone(sgd_clf)
X_train_folds = X_train[train_index]
y_train_folds = (y_train_5[train_index])
X_test_fold = X_train[test_index]
y_test_fold = (y_train_5[test_index])
clone_clf.fit(X_train_folds, y_train_folds)
y_pred = clone_clf.predict(X_test_fold)
n_correct = sum(y_pred == y_test_fold)
print(n_correct / len(y_pred))
0.9502 0.96565 0.96495
from sklearn.base import BaseEstimator
class Never5Classifier(BaseEstimator):
def fit(self, X, y=None):
pass
def predict(self, X):
return np.zeros((len(X), 1), dtype=bool)
never_5_clf = Never5Classifier()
cross_val_score(never_5_clf, X_train, y_train_5, cv=3, scoring="accuracy")
array([ 0.909 , 0.90715, 0.9128 ])
from sklearn.model_selection import cross_val_predict
y_train_pred = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3)
from sklearn.metrics import confusion_matrix
confusion_matrix(y_train_5, y_train_pred)
array([[53272, 1307], [ 1077, 4344]])
y_train_perfect_predictions = y_train_5
confusion_matrix(y_train_5, y_train_perfect_predictions)
array([[54579, 0], [ 0, 5421]])
from sklearn.metrics import precision_score, recall_score
precision_score(y_train_5, y_train_pred)
0.76871350203503808
4344 / (4344 + 1307)
0.7687135020350381
recall_score(y_train_5, y_train_pred)
0.80132816823464303
4344 / (4344 + 1077)
0.801328168234643
from sklearn.metrics import f1_score
f1_score(y_train_5, y_train_pred)
0.78468208092485547
4344 / (4344 + (1077 + 1307)/2)
0.7846820809248555
y_scores = sgd_clf.decision_function([some_digit])
y_scores
array([ 161855.74572176])
threshold = 0
y_some_digit_pred = (y_scores > threshold)
y_some_digit_pred
array([ True], dtype=bool)
threshold = 200000
y_some_digit_pred = (y_scores > threshold)
y_some_digit_pred
array([False], dtype=bool)
y_scores = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3,
method="decision_function")
Note: there is an issue introduced in Scikit-Learn 0.19.0 where the result of cross_val_predict()
is incorrect in the binary classification case when using method="decision_function"
, as in the code above. The resulting array has an extra first dimension full of 0s. We need to add this small hack for now to work around this issue:
y_scores.shape
(60000, 2)
# hack to work around issue #9589 introduced in Scikit-Learn 0.19.0
if y_scores.ndim == 2:
y_scores = y_scores[:, 1]
from sklearn.metrics import precision_recall_curve
precisions, recalls, thresholds = precision_recall_curve(y_train_5, y_scores)
def plot_precision_recall_vs_threshold(precisions, recalls, thresholds):
plt.plot(thresholds, precisions[:-1], "b--", label="Precision", linewidth=2)
plt.plot(thresholds, recalls[:-1], "g-", label="Recall", linewidth=2)
plt.xlabel("Threshold", fontsize=16)
plt.legend(loc="upper left", fontsize=16)
plt.ylim([0, 1])
plt.figure(figsize=(8, 4))
plot_precision_recall_vs_threshold(precisions, recalls, thresholds)
plt.xlim([-700000, 700000])
save_fig("precision_recall_vs_threshold_plot")
plt.show()
Saving figure precision_recall_vs_threshold_plot
(y_train_pred == (y_scores > 0)).all()
True
y_train_pred_90 = (y_scores > 70000)
precision_score(y_train_5, y_train_pred_90)
0.86592051164915484
recall_score(y_train_5, y_train_pred_90)
0.69931746910164172
def plot_precision_vs_recall(precisions, recalls):
plt.plot(recalls, precisions, "b-", linewidth=2)
plt.xlabel("Recall", fontsize=16)
plt.ylabel("Precision", fontsize=16)
plt.axis([0, 1, 0, 1])
plt.figure(figsize=(8, 6))
plot_precision_vs_recall(precisions, recalls)
save_fig("precision_vs_recall_plot")
plt.show()
Saving figure precision_vs_recall_plot
from sklearn.metrics import roc_curve
fpr, tpr, thresholds = roc_curve(y_train_5, y_scores)
def plot_roc_curve(fpr, tpr, label=None):
plt.plot(fpr, tpr, linewidth=2, label=label)
plt.plot([0, 1], [0, 1], 'k--')
plt.axis([0, 1, 0, 1])
plt.xlabel('False Positive Rate', fontsize=16)
plt.ylabel('True Positive Rate', fontsize=16)
plt.figure(figsize=(8, 6))
plot_roc_curve(fpr, tpr)
save_fig("roc_curve_plot")
plt.show()
Saving figure roc_curve_plot
from sklearn.metrics import roc_auc_score
roc_auc_score(y_train_5, y_scores)
0.96244965559671547
from sklearn.ensemble import RandomForestClassifier
forest_clf = RandomForestClassifier(random_state=42)
y_probas_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3,
method="predict_proba")
y_scores_forest = y_probas_forest[:, 1] # score = proba of positive class
fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5,y_scores_forest)
plt.figure(figsize=(8, 6))
plt.plot(fpr, tpr, "b:", linewidth=2, label="SGD")
plot_roc_curve(fpr_forest, tpr_forest, "Random Forest")
plt.legend(loc="lower right", fontsize=16)
save_fig("roc_curve_comparison_plot")
plt.show()
Saving figure roc_curve_comparison_plot
roc_auc_score(y_train_5, y_scores_forest)
0.99312433660038291
y_train_pred_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3)
precision_score(y_train_5, y_train_pred_forest)
0.98529734474434938
recall_score(y_train_5, y_train_pred_forest)
0.82826046854823832
sgd_clf.fit(X_train, y_train)
sgd_clf.predict([some_digit])
array([ 5.])
some_digit_scores = sgd_clf.decision_function([some_digit])
some_digit_scores
array([[-311402.62954431, -363517.28355739, -446449.5306454 , -183226.61023518, -414337.15339485, 161855.74572176, -452576.39616343, -471957.14962573, -518542.33997148, -536774.63961222]])
np.argmax(some_digit_scores)
5
sgd_clf.classes_
array([ 0., 1., 2., 3., 4., 5., 6., 7., 8., 9.])
sgd_clf.classes_[5]
5.0
from sklearn.multiclass import OneVsOneClassifier
ovo_clf = OneVsOneClassifier(SGDClassifier(random_state=42))
ovo_clf.fit(X_train, y_train)
ovo_clf.predict([some_digit])
/home/ageron/.virtualenvs/ml/lib/python3.5/site-packages/sklearn/linear_model/stochastic_gradient.py:84: FutureWarning: max_iter and tol parameters have been added in <class 'sklearn.linear_model.stochastic_gradient.SGDClassifier'> in 0.19. If both are left unset, they default to max_iter=5 and tol=None. If tol is not None, max_iter defaults to max_iter=1000. From 0.21, default max_iter will be 1000, and default tol will be 1e-3. "and default tol will be 1e-3." % type(self), FutureWarning)
array([ 5.])
len(ovo_clf.estimators_)
45
forest_clf.fit(X_train, y_train)
forest_clf.predict([some_digit])
array([ 5.])
forest_clf.predict_proba([some_digit])
array([[ 0.1, 0. , 0. , 0.1, 0. , 0.8, 0. , 0. , 0. , 0. ]])
cross_val_score(sgd_clf, X_train, y_train, cv=3, scoring="accuracy")
array([ 0.84063187, 0.84899245, 0.86652998])
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train.astype(np.float64))
cross_val_score(sgd_clf, X_train_scaled, y_train, cv=3, scoring="accuracy")
array([ 0.91011798, 0.90874544, 0.906636 ])
y_train_pred = cross_val_predict(sgd_clf, X_train_scaled, y_train, cv=3)
conf_mx = confusion_matrix(y_train, y_train_pred)
conf_mx
array([[5725, 3, 24, 9, 10, 49, 50, 10, 39, 4], [ 2, 6493, 43, 25, 7, 40, 5, 10, 109, 8], [ 51, 41, 5321, 104, 89, 26, 87, 60, 166, 13], [ 47, 46, 141, 5342, 1, 231, 40, 50, 141, 92], [ 19, 29, 41, 10, 5366, 9, 56, 37, 86, 189], [ 73, 45, 36, 193, 64, 4582, 111, 30, 193, 94], [ 29, 34, 44, 2, 42, 85, 5627, 10, 45, 0], [ 25, 24, 74, 32, 54, 12, 6, 5787, 15, 236], [ 52, 161, 73, 156, 10, 163, 61, 25, 5027, 123], [ 43, 35, 26, 92, 178, 28, 2, 223, 82, 5240]])
def plot_confusion_matrix(matrix):
"""If you prefer color and a colorbar"""
fig = plt.figure(figsize=(8,8))
ax = fig.add_subplot(111)
cax = ax.matshow(matrix)
fig.colorbar(cax)
plt.matshow(conf_mx, cmap=plt.cm.gray)
save_fig("confusion_matrix_plot", tight_layout=False)
plt.show()
Saving figure confusion_matrix_plot
row_sums = conf_mx.sum(axis=1, keepdims=True)
norm_conf_mx = conf_mx / row_sums
np.fill_diagonal(norm_conf_mx, 0)
plt.matshow(norm_conf_mx, cmap=plt.cm.gray)
save_fig("confusion_matrix_errors_plot", tight_layout=False)
plt.show()
Saving figure confusion_matrix_errors_plot
cl_a, cl_b = 3, 5
X_aa = X_train[(y_train == cl_a) & (y_train_pred == cl_a)]
X_ab = X_train[(y_train == cl_a) & (y_train_pred == cl_b)]
X_ba = X_train[(y_train == cl_b) & (y_train_pred == cl_a)]
X_bb = X_train[(y_train == cl_b) & (y_train_pred == cl_b)]
plt.figure(figsize=(8,8))
plt.subplot(221); plot_digits(X_aa[:25], images_per_row=5)
plt.subplot(222); plot_digits(X_ab[:25], images_per_row=5)
plt.subplot(223); plot_digits(X_ba[:25], images_per_row=5)
plt.subplot(224); plot_digits(X_bb[:25], images_per_row=5)
save_fig("error_analysis_digits_plot")
plt.show()
Saving figure error_analysis_digits_plot
from sklearn.neighbors import KNeighborsClassifier
y_train_large = (y_train >= 7)
y_train_odd = (y_train % 2 == 1)
y_multilabel = np.c_[y_train_large, y_train_odd]
knn_clf = KNeighborsClassifier()
knn_clf.fit(X_train, y_multilabel)
KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski', metric_params=None, n_jobs=1, n_neighbors=5, p=2, weights='uniform')
knn_clf.predict([some_digit])
array([[False, True]], dtype=bool)
Warning: the following cell may take a very long time (possibly hours depending on your hardware).
y_train_knn_pred = cross_val_predict(knn_clf, X_train, y_multilabel, cv=3)
f1_score(y_multilabel, y_train_knn_pred, average="macro")
0.97709078477525002
noise = np.random.randint(0, 100, (len(X_train), 784))
X_train_mod = X_train + noise
noise = np.random.randint(0, 100, (len(X_test), 784))
X_test_mod = X_test + noise
y_train_mod = X_train
y_test_mod = X_test
some_index = 5500
plt.subplot(121); plot_digit(X_test_mod[some_index])
plt.subplot(122); plot_digit(y_test_mod[some_index])
save_fig("noisy_digit_example_plot")
plt.show()
Saving figure noisy_digit_example_plot
knn_clf.fit(X_train_mod, y_train_mod)
clean_digit = knn_clf.predict([X_test_mod[some_index]])
plot_digit(clean_digit)
save_fig("cleaned_digit_example_plot")
Saving figure cleaned_digit_example_plot
from sklearn.dummy import DummyClassifier
dmy_clf = DummyClassifier()
y_probas_dmy = cross_val_predict(dmy_clf, X_train, y_train_5, cv=3, method="predict_proba")
y_scores_dmy = y_probas_dmy[:, 1]
fprr, tprr, thresholdsr = roc_curve(y_train_5, y_scores_dmy)
plot_roc_curve(fprr, tprr)
from sklearn.neighbors import KNeighborsClassifier
knn_clf = KNeighborsClassifier(n_jobs=-1, weights='distance', n_neighbors=4)
knn_clf.fit(X_train, y_train)
KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski', metric_params=None, n_jobs=-1, n_neighbors=4, p=2, weights='distance')
y_knn_pred = knn_clf.predict(X_test)
from sklearn.metrics import accuracy_score
accuracy_score(y_test, y_knn_pred)
0.97140000000000004
from scipy.ndimage.interpolation import shift
def shift_digit(digit_array, dx, dy, new=0):
return shift(digit_array.reshape(28, 28), [dy, dx], cval=new).reshape(784)
plot_digit(shift_digit(some_digit, 5, 1, new=100))
X_train_expanded = [X_train]
y_train_expanded = [y_train]
for dx, dy in ((1, 0), (-1, 0), (0, 1), (0, -1)):
shifted_images = np.apply_along_axis(shift_digit, axis=1, arr=X_train, dx=dx, dy=dy)
X_train_expanded.append(shifted_images)
y_train_expanded.append(y_train)
X_train_expanded = np.concatenate(X_train_expanded)
y_train_expanded = np.concatenate(y_train_expanded)
X_train_expanded.shape, y_train_expanded.shape
((300000, 784), (300000,))
knn_clf.fit(X_train_expanded, y_train_expanded)
KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski', metric_params=None, n_jobs=-1, n_neighbors=4, p=2, weights='distance')
y_knn_expanded_pred = knn_clf.predict(X_test)
accuracy_score(y_test, y_knn_expanded_pred)
0.97629999999999995
ambiguous_digit = X_test[2589]
knn_clf.predict_proba([ambiguous_digit])
array([[ 0. , 0. , 0.5053645, 0. , 0. , 0. , 0. , 0.4946355, 0. , 0. ]])
plot_digit(ambiguous_digit)
Coming soon