文章目錄

• 轉(zhuǎn)換器與估計器
• 分類算法-K近鄰算法
• • 一個例子弄懂k-近鄰
  • 計算距離公式
  • sklearn.neighbors
  • Method
  • k近鄰實(shí)例
  • k-近鄰算法案例分析
  • 對Iris數(shù)據(jù)集進(jìn)行分割
  • 對特征數(shù)據(jù)進(jìn)行標(biāo)準(zhǔn)化
• 樸素貝葉斯
• • 概率論基礎(chǔ)
  • 聯(lián)合概率與條件概率
  • 聯(lián)合概率
  • 條件概率
  • 如果每個事件相互獨(dú)立
  • 拉普拉斯平滑
  • sklearn樸素貝葉斯實(shí)現(xiàn)API
  • 樸素貝葉斯優(yōu)缺點(diǎn)
  • 詞袋法特征值計算
  • 案例

轉(zhuǎn)換器與估計器

分類算法-K近鄰算法

一個例子弄懂k-近鄰

電影可以按照題材分類，每個題材又是如何定義的呢？那么假如兩種類型的電影，動作片和愛情片。動作片有哪些公共的特征？那么愛情片又存在哪些明顯的差別呢？我們發(fā)現(xiàn)動作片中打斗鏡頭的次數(shù)較多，而愛情片中接吻鏡頭相對更多。當(dāng)然動作片中也有一些接吻鏡頭，愛情片中也會有一些打斗鏡頭。所以不能單純通過是否存在打斗鏡頭或者接吻鏡頭來判斷影片的類別。那么現(xiàn)在我們有6部影片已經(jīng)明確了類別，也有打斗鏡頭和接吻鏡頭的次數(shù)，還有一部電影類型未知

那么我們使用K-近鄰算法來分類愛情片和動作片：存在一個樣本數(shù)據(jù)集合，也叫訓(xùn)練樣本集，樣本個數(shù)M個，知道每一個數(shù)據(jù)特征與類別對應(yīng)關(guān)系，然后存在未知類型數(shù)據(jù)集合1個，那么我們要選擇一個測試樣本數(shù)據(jù)中與訓(xùn)練樣本中M個的距離，排序過后選出最近的K個，這個取值一般不大于20個。選擇K個最相近數(shù)據(jù)中次數(shù)最多的分類。那么我們根據(jù)這個原則去判斷未知電影的分類

我們假設(shè)K為3，那么排名前三個電影的類型都是愛情片，所以我們判定這個未知電影也是一個愛情片。那么計算距離是怎樣計算的呢？
歐氏距離 那么對于兩個向量點(diǎn)

??之間的距離,可以通過該公式表示

計算距離公式

歐氏距離
k近鄰算法需要做標(biāo)準(zhǔn)化處理

sklearn.neighbors

sklearn.neighbors提供監(jiān)督的基于鄰居的學(xué)習(xí)方法的功能，sklearn.neighbors.KNeighborsClassifier是一個最近鄰居分類器。那么KNeighborsClassifier是一個類，我們看一下實(shí)例化時候的參數(shù)

class sklearn.neighbors.KNeighborsClassifier(n_neighbors=5, weights='uniform', algorithm='auto', leaf_size=30, p=2, metric='minkowski', metric_params=None, n_jobs=1, **kwargs)**""":param n_neighbors：int，可選（默認(rèn)= 5），k_neighbors查詢默認(rèn)使用的鄰居數(shù):param algorithm：{'auto'，'ball_tree'，'kd_tree'，'brute'}，可選用于計算最近鄰居的算法：'ball_tree'將會使用 BallTree，'kd_tree'將使用 KDTree，“野獸”將使用強(qiáng)力搜索。'auto'將嘗試根據(jù)傳遞給fit方法的值來決定最合適的算法。:param n_jobs：int，可選（默認(rèn)= 1),用于鄰居搜索的并行作業(yè)數(shù)。如果-1，則將作業(yè)數(shù)設(shè)置為CPU內(nèi)核數(shù)。不影響fit方法。""" import numpy as np from sklearn.neighbors import KNeighborsClassifierneigh = KNeighborsClassifier(n_neighbors=3)

Method

fit(X, y)
使用X作為訓(xùn)練數(shù)據(jù)擬合模型，y作為X的類別值。X，y為數(shù)組或者矩陣

X = np.array([[1,1],[1,1.1],[0,0],[0,0.1]]) y = np.array([1,1,0,0]) neigh.fit(X,y)

kneighbors(X=None, n_neighbors=None, return_distance=True)
找到指定點(diǎn)集X的n_neighbors個鄰居，return_distance為False的話，不返回距離

neigh.kneighbors(np.array([[1.1,1.1]]),return_distance= False)neigh.kneighbors(np.array([[1.1,1.1]]),return_distance= False,n_neighbors=2)

predict(X)
預(yù)測提供的數(shù)據(jù)的類標(biāo)簽

neigh.predict(np.array([[0.1,0.1],[1.1,1.1]]))

predict_proba(X)
返回測試數(shù)據(jù)X屬于某一類別的概率估計

neigh.predict_proba(np.array([[1.1,1.1]]))

k近鄰實(shí)例

def knncls():"""K-近鄰預(yù)測用戶簽到位置:return:None"""# 讀取數(shù)據(jù)data = pd.read_csv("./data/FBlocation/train.csv")# print(data.head(10))# 處理數(shù)據(jù)# 1、縮小數(shù)據(jù),查詢數(shù)據(jù)曬訊data = data.query("x > 1.0 & x < 1.25 & y > 2.5 & y < 2.75")# 處理時間的數(shù)據(jù)time_value = pd.to_datetime(data['time'], unit='s')print(time_value)# 把日期格式轉(zhuǎn)換成 字典格式time_value = pd.DatetimeIndex(time_value)# 構(gòu)造一些特征data['day'] = time_value.daydata['hour'] = time_value.hourdata['weekday'] = time_value.weekday# 把時間戳特征刪除data = data.drop(['time'], axis=1)print(data)# 把簽到數(shù)量少于n個目標(biāo)位置刪除place_count = data.groupby('place_id').count()tf = place_count[place_count.row_id > 3].reset_index()data = data[data['place_id'].isin(tf.place_id)]# 取出數(shù)據(jù)當(dāng)中的特征值和目標(biāo)值y = data['place_id']x = data.drop(['place_id'], axis=1)# 進(jìn)行數(shù)據(jù)的分割訓(xùn)練集合測試集x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.25)# 特征工程（標(biāo)準(zhǔn)化）std = StandardScaler()# 對測試集和訓(xùn)練集的特征值進(jìn)行標(biāo)準(zhǔn)化x_train = std.fit_transform(x_train)x_test = std.transform(x_test)# 進(jìn)行算法流程 # 超參數(shù)knn = KNeighborsClassifier()# # fit， predict,score# knn.fit(x_train, y_train)## # 得出預(yù)測結(jié)果# y_predict = knn.predict(x_test)## print("預(yù)測的目標(biāo)簽到位置為：", y_predict)## # 得出準(zhǔn)確率# print("預(yù)測的準(zhǔn)確率:", knn.score(x_test, y_test))# 構(gòu)造一些參數(shù)的值進(jìn)行搜索param = {"n_neighbors": [3, 5, 10]}# 進(jìn)行網(wǎng)格搜索gc = GridSearchCV(knn, param_grid=param, cv=2)gc.fit(x_train, y_train)# 預(yù)測準(zhǔn)確率print("在測試集上準(zhǔn)確率：", gc.score(x_test, y_test))print("在交叉驗(yàn)證當(dāng)中最好的結(jié)果：", gc.best_score_)print("選擇最好的模型是：", gc.best_estimator_)print("每個超參數(shù)每次交叉驗(yàn)證的結(jié)果：", gc.cv_results_)return None C:\Users\HP\Anaconda3\python.exe D:/PycharmProjects/untitled2/算法/算法3.pyrow_id x y accuracy time place_id 0 0 0.7941 9.0809 54 470702 8523065625 1 1 5.9567 4.7968 13 186555 1757726713 2 2 8.3078 7.0407 74 322648 1137537235 3 3 7.3665 2.5165 65 704587 6567393236 4 4 4.0961 1.1307 31 472130 7440663949 5 5 3.8099 1.9586 75 178065 6289802927 6 6 6.3336 4.3720 13 666829 9931249544 7 7 5.7409 6.7697 85 369002 5662813655 8 8 4.3114 6.9410 3 166384 8471780938 9 9 6.3414 0.0758 65 400060 1253803156 600 1970-01-01 18:09:40 957 1970-01-10 02:11:10 4345 1970-01-05 15:08:02 4735 1970-01-06 23:03:03 5580 1970-01-09 11:26:50 6090 1970-01-02 16:25:07 6234 1970-01-04 15:52:57 6350 1970-01-01 10:13:36 7468 1970-01-09 15:26:06 8478 1970-01-08 23:52:02 9357 1970-01-04 16:53:19 12125 1970-01-07 03:55:07 14937 1970-01-06 03:46:38 20660 1970-01-08 03:08:15 20930 1970-01-02 21:31:48 21731 1970-01-07 08:52:19 26584 1970-01-04 15:48:09 27937 1970-01-08 03:51:54 30798 1970-01-01 20:58:30 33184 1970-01-06 15:31:39 33877 1970-01-02 14:58:01 34340 1970-01-04 14:03:40 37405 1970-01-04 15:35:01 38968 1970-01-08 08:56:00 41861 1970-01-01 03:13:36 42135 1970-01-02 02:36:41 42729 1970-01-01 16:03:37 44283 1970-01-08 06:48:09 44549 1970-01-07 01:10:01 44694 1970-01-08 14:30:07... 29070221 1970-01-07 02:55:07 29070322 1970-01-01 18:13:24 29070934 1970-01-03 03:44:08 29071712 1970-01-08 04:19:17 29072165 1970-01-04 12:42:07 29073572 1970-01-07 20:29:38 29074121 1970-01-08 02:30:21 29077579 1970-01-08 18:08:30 29077716 1970-01-09 11:51:11 29079070 1970-01-07 00:33:24 29079416 1970-01-05 10:48:15 29079931 1970-01-02 05:35:45 29083241 1970-01-02 00:35:10 29083789 1970-01-05 09:39:49 29084739 1970-01-06 12:04:17 29085497 1970-01-03 11:31:33 29086167 1970-01-08 01:04:37 29087094 1970-01-04 22:25:01 29089004 1970-01-01 23:26:24 29090443 1970-01-03 09:00:22 29093677 1970-01-07 10:03:36 29094547 1970-01-09 11:44:34 29096155 1970-01-04 08:07:44 29099420 1970-01-04 15:47:47 29099686 1970-01-08 01:24:11 29100203 1970-01-01 10:33:56 29108443 1970-01-07 23:22:04 29109993 1970-01-08 15:03:14 29111539 1970-01-04 00:53:41 29112154 1970-01-08 23:01:07 Name: time, Length: 17710, dtype: datetime64[ns] D:/PycharmProjects/untitled2/算法/算法3.py:36: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value insteadSee the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copydata['day']=time_value.day D:/PycharmProjects/untitled2/算法/算法3.py:37: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value insteadSee the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copydata['hour']=time_value.hour D:/PycharmProjects/untitled2/算法/算法3.py:38: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value insteadSee the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copydata['weekday']=time_value.weekdayrow_id x y accuracy place_id day hour weekday 600 600 1.2214 2.7023 17 6683426742 1 18 3 957 957 1.1832 2.6891 58 6683426742 10 2 5 4345 4345 1.1935 2.6550 11 6889790653 5 15 0 4735 4735 1.1452 2.6074 49 6822359752 6 23 1 5580 5580 1.0089 2.7287 19 1527921905 9 11 4 6090 6090 1.1140 2.6262 11 4000153867 2 16 4 6234 6234 1.1449 2.5003 34 3741484405 4 15 6 6350 6350 1.0844 2.7436 65 5963693798 1 10 3 7468 7468 1.0058 2.5096 66 9076695703 9 15 4 8478 8478 1.2015 2.5187 72 3992589015 8 23 3 9357 9357 1.1916 2.7323 170 5163401947 4 16 6 12125 12125 1.1388 2.5029 69 7536975002 7 3 2 14937 14937 1.1426 2.7441 11 6780386626 6 3 1 20660 20660 1.2387 2.5959 65 3683087833 8 3 3 20930 20930 1.0519 2.5208 67 6399991653 2 21 4 21731 21731 1.2171 2.7263 99 8048985799 7 8 2 26584 26584 1.1235 2.6282 63 5606572086 4 15 6 27937 27937 1.1287 2.6332 588 5606572086 8 3 3 30798 30798 1.0422 2.6474 49 1435128522 1 20 3 33184 33184 1.0128 2.5865 75 1913341282 6 15 1 33877 33877 1.1437 2.6972 972 6683426742 2 14 4 34340 34340 1.1513 2.5824 176 2355236719 4 14 6 37405 37405 1.2122 2.7106 10 2946102544 4 15 6 38968 38968 1.1496 2.6298 166 9598377925 8 8 3 41861 41861 1.0886 2.6840 10 3312463746 1 3 3 42135 42135 1.0498 2.6840 5 3312463746 2 2 4 42729 42729 1.0694 2.5829 10 1812226671 1 16 3 44283 44283 1.2384 2.7398 60 8048985799 8 6 3 44549 44549 1.2077 2.5370 76 3992589015 7 1 2 44694 44694 1.0380 2.5315 152 5035268417 8 14 3 ... ... ... ... ... ... ... ... ... 29070221 29070221 1.1678 2.5605 66 2355236719 7 2 2 29070322 29070322 1.0493 2.7010 74 3312463746 1 18 3 29070934 29070934 1.1899 2.5176 28 2199223958 3 3 5 29071712 29071712 1.2260 2.7367 4 2946102544 8 4 3 29072165 29072165 1.0175 2.6220 42 5283227804 4 12 6 29073572 29073572 1.2467 2.7316 64 8048985799 7 20 2 29074121 29074121 1.2071 2.6646 161 5270522918 8 2 3 29077579 29077579 1.2479 2.6474 42 2006503124 8 18 3 29077716 29077716 1.1898 2.7013 5 6683426742 9 11 4 29079070 29079070 1.1882 2.5476 28 1731306153 7 0 2 29079416 29079416 1.2335 2.5903 72 6766324666 5 10 0 29079931 29079931 1.0213 2.6554 167 5270522918 2 5 4 29083241 29083241 1.0600 2.6722 71 9632980559 2 0 4 29083789 29083789 1.0674 2.6184 88 1097200869 5 9 0 29084739 29084739 1.2319 2.6767 63 2327054745 6 12 1 29085497 29085497 1.0550 2.5997 175 1097200869 3 11 5 29086167 29086167 1.0515 2.6758 57 6237569496 8 1 3 29087094 29087094 1.0088 2.5978 71 1097200869 4 22 6 29089004 29089004 1.1860 2.6926 153 2215268322 1 23 3 29090443 29090443 1.0568 2.6959 58 2460093296 3 9 5 29093677 29093677 1.0016 2.5252 16 9013153173 7 10 2 29094547 29094547 1.1101 2.6530 24 5270522918 9 11 4 29096155 29096155 1.0122 2.6450 65 8178619377 4 8 6 29099420 29099420 1.1675 2.5556 9 2355236719 4 15 6 29099686 29099686 1.0405 2.6723 13 3312463746 8 1 3 29100203 29100203 1.0129 2.6775 12 3312463746 1 10 3 29108443 29108443 1.1474 2.6840 36 3533177779 7 23 2 29109993 29109993 1.0240 2.7238 62 6424972551 8 15 3 29111539 29111539 1.2032 2.6796 87 3533177779 4 0 6 29112154 29112154 1.1070 2.5419 178 4932578245 8 23 3[17710 rows x 8 columns] C:\Users\HP\Anaconda3\lib\site-packages\sklearn\preprocessing\data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.return self.partial_fit(X, y) C:\Users\HP\Anaconda3\lib\site-packages\sklearn\base.py:464: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.return self.fit(X, **fit_params).transform(X) D:/PycharmProjects/untitled2/算法/算法3.py:62: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.x_test=std.transform(x_test) C:\Users\HP\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py:652: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=2.% (min_groups, self.n_splits)), Warning) 在測試集上準(zhǔn)確率 0.42174940898345153 在交叉驗(yàn)證中最好結(jié)果 0.3899747793190416 選擇最好的模型 KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',metric_params=None, n_jobs=None, n_neighbors=10, p=2,weights='uniform') 每個超參數(shù)每次交叉驗(yàn)證的結(jié)果 {'mean_fit_time': array([0.01645494, 0.01138353, 0.01047194]), 'std_fit_time': array([0.00448656, 0.00058341, 0.00249255]), 'mean_score_time': array([0.75878692, 0.61585569, 0.6273967 ]), 'std_score_time': array([0.01371908, 0.09724212, 0.01204288]), 'param_n_neighbors': masked_array(data=[3, 5, 10],mask=[False, False, False],fill_value='?',dtype=object), 'params': [{'n_neighbors': 3}, {'n_neighbors': 5}, {'n_neighbors': 10}], 'split0_test_score': array([0.33999688, 0.36842105, 0.38841168]), 'split1_test_score': array([0.34622116, 0.37549722, 0.39156722]), 'mean_test_score': array([0.34308008, 0.37192623, 0.38997478]), 'std_test_score': array([0.00311201, 0.00353793, 0.0015777 ]), 'rank_test_score': array([3, 2, 1]), 'split0_train_score': array([0.60875099, 0.56022275, 0.49880668]), 'split1_train_score': array([0.59253475, 0.53959082, 0.48633453]), 'mean_train_score': array([0.60064287, 0.54990678, 0.49257061]), 'std_train_score': array([0.00810812, 0.01031597, 0.00623608])}Process finished with exit code 0

k-近鄰算法案例分析

本案例使用最著名的”鳶尾“數(shù)據(jù)集，該數(shù)據(jù)集曾經(jīng)被Fisher用在經(jīng)典論文中，目前作為教科書般的數(shù)據(jù)樣本預(yù)存在Scikit-learn的工具包中

from sklearn.datasets import load_iris # 使用加載器讀取數(shù)據(jù)并且存入變量iris iris = load_iris()# 查驗(yàn)數(shù)據(jù)規(guī)模 iris.data.shape# 查看數(shù)據(jù)說明（這是一個好習(xí)慣） print iris.DESCR

通過上述代碼對數(shù)據(jù)的查驗(yàn)以及數(shù)據(jù)本身的描述，我們了解到Iris數(shù)據(jù)集共有150朵鳶尾數(shù)據(jù)樣本，并且均勻分布在3個不同的亞種；每個數(shù)據(jù)樣本有總共4個不同的關(guān)于花瓣、花萼的形狀特征所描述。由于沒有制定的測試集合，因此按照慣例，我們需要對數(shù)據(jù)進(jìn)行隨即分割，25%的樣本用于測試，其余75%的樣本用于模型的訓(xùn)練。
由于不清楚數(shù)據(jù)集的排列是否隨機(jī)，可能會有按照類別去進(jìn)行依次排列，這樣訓(xùn)練樣本的不均衡的，所以我們需要分割數(shù)據(jù)，已經(jīng)默認(rèn)有隨機(jī)采樣的功能

對Iris數(shù)據(jù)集進(jìn)行分割

from sklearn.cross_validation import train_test_split X_train,X_test,y_train,y_test = train_test_split(iris.data,iris.target,test_size=0.25,random_state=42)

對特征數(shù)據(jù)進(jìn)行標(biāo)準(zhǔn)化

from sklearn.preprocessing import StandardScalerss = StandardScaler() X_train = ss.fit_transform(X_train) X_test = ss.fit_transform(X_test)

K近鄰算法是非常直觀的機(jī)器學(xué)習(xí)模型，我們可以發(fā)現(xiàn)K近鄰算法沒有參數(shù)訓(xùn)練過程，也就是說，我們沒有通過任何學(xué)習(xí)算法分析訓(xùn)練數(shù)據(jù)，而只是根據(jù)測試樣本訓(xùn)練數(shù)據(jù)的分布直接作出分類決策。因此，K近鄰屬于無參數(shù)模型中非常簡單一種。

from sklearn.datasets import load_iris from sklearn.cross_validation import train_test_split from sklearn.preprocessing import StandardScaler from sklearn.neighbors import KNeighborsClassifier from sklearn.metrics import classification_report from sklearn.model_selection import GridSearchCVdef knniris():"""鳶尾花分類:return: None"""# 數(shù)據(jù)集獲取和分割lr = load_iris()x_train, x_test, y_train, y_test = train_test_split(lr.data, lr.target, test_size=0.25)# 進(jìn)行標(biāo)準(zhǔn)化std = StandardScaler()x_train = std.fit_transform(x_train)x_test = std.transform(x_test)# estimator流程knn = KNeighborsClassifier()# # 得出模型# knn.fit(x_train,y_train)## # 進(jìn)行預(yù)測或者得出精度# y_predict = knn.predict(x_test)## # score = knn.score(x_test,y_test)# 通過網(wǎng)格搜索,n_neighbors為參數(shù)列表param = {"n_neighbors": [3, 5, 7]}gs = GridSearchCV(knn, param_grid=param, cv=10)# 建立模型gs.fit(x_train,y_train)# print(gs)# 預(yù)測數(shù)據(jù)print(gs.score(x_test,y_test))# 分類模型的精確率和召回率# print("每個類別的精確率與召回率：",classification_report(y_test, y_predict,target_names=lr.target_names))return Noneif __name__ == "__main__":knniris()

樸素貝葉斯

樸素貝葉斯（Naive Bayes）是一個非常簡單，但是實(shí)用性很強(qiáng)的分類模型。樸素貝葉斯分類器的構(gòu)造基礎(chǔ)是貝葉斯理論。

概率論基礎(chǔ)

概率定義為一件事情發(fā)生的可能性。事情發(fā)生的概率可以 通過觀測數(shù)據(jù)中的事件發(fā)生次數(shù)來計算，事件發(fā)生的概率等于改事件發(fā)生次數(shù)除以所有事件發(fā)生的總次數(shù)。舉一些例子：
扔出一個硬幣，結(jié)果頭像朝上
某天是晴天
某個單詞在未知文檔中出現(xiàn)

聯(lián)合概率與條件概率

聯(lián)合概率

是指兩件事情同時發(fā)生的概率。那么我們假設(shè)樣本空間有一些天氣數(shù)據(jù)：

條件概率

如果每個事件相互獨(dú)立

拉普拉斯平滑

sklearn樸素貝葉斯實(shí)現(xiàn)API

class sklearn.naive_bayes.MultinomialNB(alpha=1.0, fit_prior=True, class_prior=None)""":param alpha：float，optional（default = 1.0）加法（拉普拉斯/ Lidstone）平滑參數(shù)（0為無平滑）"""

樸素貝葉斯優(yōu)缺點(diǎn)

詞袋法特征值計算

案例

互聯(lián)網(wǎng)新聞分類

def naviebayes():"""樸素貝葉斯進(jìn)行文本分類:return: None"""news=fetch_20newsgroups(subset='all')#進(jìn)行數(shù)據(jù)分割x_train,x_test,y_train,y_test=train_test_split(news.data,news.target,test_size=0.25)#對數(shù)據(jù)集進(jìn)行特征抽取tf=TfidfVectorizer()# 以訓(xùn)練集當(dāng)中的詞的列表進(jìn)行每篇文章重要性統(tǒng)計['a','b','c','d']x_train=tf.fit_transform(x_train)# print(tf.get_feature_names())x_test=tf.transform(x_test)#進(jìn)行樸素貝葉斯預(yù)測mlt=MultinomialNB(alpha=1.0)# print(x_train.toarray())mlt.fit(x_train,y_train)y_predict=mlt.predict(x_test)print("預(yù)測文章類別",y_predict)#得出準(zhǔn)確率# print("準(zhǔn)確率為",mlt.score(x_test,y_test))# print("每個類別準(zhǔn)確率和召回率",classification_report(y_test,y_predict,target_names=news.target_names))return None

總結(jié)

以上是生活随笔為你收集整理的机器学习算法3的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。