本文是對(duì)書《機(jī)器學(xué)習(xí)實(shí)戰(zhàn)：基于Scikit-Learn和Tensorflow》第三章的知識(shí)學(xué)習(xí)以及代碼復(fù)現(xiàn)，歡迎大家一起學(xué)習(xí)一起進(jìn)步。

獲取數(shù)據(jù)集

提前將MNIST數(shù)據(jù)集下載好，并放在'\scikit_learn_data'目錄之下

from sklearn.datasets import fetch_mldata
mnist = fetch_mldata('MNIST original', data_home=r'C:\Users\12637\scikit_learn_data')
mnist

001.png

X, y = mnist["data"], mnist["target"]
X.shape

002.png

一共有70000張圖片，每張圖片有784個(gè)特征。因?yàn)閳D片是28x28像素，每個(gè)特征代表了一個(gè)像素點(diǎn)的強(qiáng)度，從0（白色）到255（黑色）。隨手抓取一個(gè)實(shí)例的特征向量，將其重新形成一個(gè)28X28數(shù)組，然后用Matplotlib的imshow()函數(shù)將其顯示出來

%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")
plt.show()

003.png

# MNIST數(shù)據(jù)集中的部分?jǐn)?shù)字圖像
X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]

# 給訓(xùn)練集數(shù)據(jù)洗牌
import numpy as np

shuffle_index = np.random.permutation(60000)
X_train, y_train = X_train[shuffle_index], y_train[shuffle_index]

訓(xùn)練一個(gè)二元分類器

# 訓(xùn)練一個(gè)二元分類器
y_train_5 = (y_train == 5)         # True for all 5s, False for all other digits
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)

sgd_clf.predict([some_digit])

004.png

實(shí)施交叉驗(yàn)證

# 實(shí)施交叉驗(yàn)證

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))

005.png

006.png

混淆矩陣

# 混淆矩陣

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)

007.png

精度和召回率以及精度/召回率平衡

008.png

y_scores = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3, method="decision_function")

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")
    plt.plot(thresholds, recalls[:-1], "g-", label="Recall")
    plt.xlabel("Threshold")
    plt.legend(loc="upper left")
    plt.ylim([0, 1])
    
plot_precision_recall_vs_threshold(precisions, recalls, thresholds)
plt.show()

009.png

ROC曲線繪制

# ROC曲線
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')
    plt.ylabel('True Positive Rate')
plot_roc_curve(fpr, tpr)
plt.show()

010.png

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]
fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5, y_scores_forest)

plt.plot(fpr, tpr, "b:", label="SGD")
plot_roc_curve(fpr_forest, tpr_forest, "Random Forest")
plt.legend(loc="bottom right")
plt.show()

011.png

多類別分類器

012.png

013.png

錯(cuò)誤分析

# 錯(cuò)誤分析

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

014.png

015.png

016.png

多標(biāo)簽分類

# 多標(biāo)簽分類

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)
knn_clf.predict([some_digit])

多輸出分類

# 多輸出分類

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

def plot_digit(data):
    image = data.reshape(28, 28)
    plt.imshow(image, cmap = matplotlib.cm.binary,
               interpolation="nearest")
    plt.axis("off")
    
some_index = 5500

plt.subplot(121); plot_digit(X_test_mod[some_index])
plt.subplot(122); plot_digit(y_test_mod[some_index])
plt.show()

017.png

knn_clf.fit(X_train_mod, y_train_mod)
clean_digit = knn_clf.predict([X_test_mod[some_index]])
plot_digit(clean_digit)

018.png