Spark机器学习基础-无监督学习
0.K-means
from __future__ import print_function from pyspark.ml.clustering import KMeans#硬聚类 #from pyspark.ml.evaluation import ClusteringEvaluator#2.2版本支持评估，2.1版本不支持 from pyspark.sql import SparkSession
! head -5 data/mllib/sample_kmeans_data.txt#展示前5行
结果：
0 1:0.0 2:0.0 3:0.0 1 1:0.1 2:0.1 3:0.1 2 1:0.2 2:0.2 3:0.2 3 1:9.0 2:9.0 3:9.0 4 1:9.1 2:9.1 3:9.1
spark = SparkSession\
.builder\
.appName("KMeansExample")\
.getOrCreate()
dataset = spark.read.format("libsvm").load("data/mllib/sample_kmeans_data.txt")#libsvm主要用于保存稀疏数据
# 训练K-means聚类模型
kmeans = KMeans().setK(2).setSeed(1)#setK设定聚类中心个数
model = kmeans.fit(dataset)
# 预测(即分配聚类中心)
predictions = model.transform(dataset)
# 根据Silhouette得分评估(pyspark2.2里新加)
#evaluator = ClusteringEvaluator()
#silhouette = evaluator.evaluate(predictions)
#print("Silhouette with squared euclidean distance = " + str(silhouette))
# 输出预测结果
print("predicted Center: ")
for center in predictions[['prediction']].collect():
print(center.asDict())
# 聚类中心
centers = model.clusterCenters()
print("Cluster Centers: ")
for center in centers:
print(center)
spark.stop()
结果：
predicted Center:
{'prediction': 0}
{'prediction': 0}
{'prediction': 0}
{'prediction': 1}
{'prediction': 1}
{'prediction': 1}
Cluster Centers:
[ 0.1 0.1 0.1]
[ 9.1 9.1 9.1]
2.GMM模型
from __future__ import print_function from pyspark.ml.clustering import GaussianMixture#软聚类，可以看看和KMeans的区别 from pyspark.sql import SparkSession
spark = SparkSession\
.builder\
.appName("GaussianMixtureExample")\
.getOrCreate()
dataset = spark.read.format("libsvm").load("data/mllib/sample_kmeans_data.txt")
gmm = GaussianMixture().setK(2).setSeed(0)#setK(2)设定两个高斯成分，不同的均值和方差，利用不同的权重去拟合
model = gmm.fit(dataset)
print("Gaussians shown as a DataFrame: ")
model.gaussiansDF.show(truncate=False)
spark.stop()
结果：
#pyspark 2.2写法
spark = SparkSession\
.builder\
.appName("FPGrowthExample")\
.getOrCreate()
df = spark.createDataFrame([
(0, [1, 2, 5]),
(1, [1, 2, 3, 5]),
(2, [1, 2])
], ["id", "items"])
fpGrowth = FPGrowth(itemsCol="items", minSupport=0.5, minConfidence=0.6)
model = fpGrowth.fit(df)
# Display frequent itemsets.
model.freqItemsets.show()
# Display generated association rules.
model.associationRules.show()
# transform examines the input items against all the association rules and summarize the
# consequents as prediction
model.transform(df).show()
spark.stop()
4.LDA主题模型
from __future__ import print_function from pyspark.ml.clustering import LDA from pyspark.sql import SparkSession
! head -5 data/mllib/sample_lda_libsvm_data.txt
结果：
0 1:1 2:2 3:6 4:0 5:2 6:3 7:1 8:1 9:0 10:0 11:3 1 1:1 2:3 3:0 4:1 5:3 6:0 7:0 8:2 9:0 10:0 11:1 2 1:1 2:4 3:1 4:0 5:0 6:4 7:9 8:0 9:1 10:2 11:0 3 1:2 2:1 3:0 4:3 5:0 6:0 7:5 8:0 9:2 10:3 11:9 4 1:3 2:1 3:1 4:9 5:3 6:0 7:2 8:0 9:0 10:1 11:3
spark = SparkSession \
.builder \
.appName("LDAExample") \
.getOrCreate()
# 加载数据
dataset = spark.read.format("libsvm").load("data/mllib/sample_lda_libsvm_data.txt")
# 训练LDA模型
lda = LDA(k=10, maxIter=10)#k=10：10个主题
model = lda.fit(dataset)
ll = model.logLikelihood(dataset)
lp = model.logPerplexity(dataset)
print("The lower bound on the log likelihood of the entire corpus: " + str(ll))
print("The upper bound on perplexity: " + str(lp)+"\n")
# 输出主题
topics = model.describeTopics(3)#这里设置的是3个词作为截断，可以多设置几个
print("The topics described by their top-weighted terms:")
topics.show(truncate=False)
# 数据集解析
print("transform dataset:\n")
transformed = model.transform(dataset)
transformed.show(truncate=False)
spark.stop()
结果：
The lower bound on the log likelihood of the entire corpus: -806.81672765 The upper bound on perplexity: 3.10314127288 The topics described by their top-weighted terms: +-----+-----------+---------------------------------------------------------------+ |topic|termIndices|termWeights | +-----+-----------+---------------------------------------------------------------+ |0 |[4, 7, 10] |[0.10782283322528141, 0.09748059064869798, 0.09623489511403283]| |1 |[1, 6, 9] |[0.16755677717574005, 0.14746677066462868, 0.12291625834665457]| |2 |[1, 3, 9] |[0.10064404373379261, 0.10044232016910744, 0.09911430786912553]| |3 |[3, 10, 4] |[0.2405485337093881, 0.11474862445349779, 0.09436360804237896] | |4 |[9, 10, 3] |[0.10479881323144603, 0.10207366164963672, 0.0981847998287497] | |5 |[8, 5, 7] |[0.10843492932441408, 0.09701504850837554, 0.09334497740169005]| |6 |[8, 5, 0] |[0.09874156843227488, 0.09654281376143092, 0.09565958598645523]| |7 |[9, 4, 7] |[0.11252485087182341, 0.09755086126590837, 0.09643430677076377]| |8 |[4, 1, 2] |[0.10994282164614115, 0.09410686880245682, 0.09374715192052394]| |9 |[5, 4, 0] |[0.1526594065996145, 0.1401540984288492, 0.13878637240223393] | +-----+-----------+---------------------------------------------------------------+ transform dataset: +-----+---------------------------------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ |label|features |topicDistribution | +-----+---------------------------------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ |0.0 |(11,[0,1,2,4,5,6,7,10],[1.0,2.0,6.0,2.0,3.0,1.0,1.0,3.0]) |[0.004830688530254547,0.9563372032839312,0.004830653288196159,0.0049247000529390305,0.0048306686997597464,0.004830691229644231,0.004830725952841193,0.0048306754566327355,0.004830728026376915,0.004923265479424051] | |1.0 |(11,[0,1,3,4,7,10],[1.0,3.0,1.0,3.0,2.0,1.0]) |[0.008057778649104173,0.3148301429775185,0.0080578223830065,0.008215777942952482,0.008057720361154553,0.008057732489412228,0.008057717726124932,0.00805779103670622,0.008057840506543925,0.6205496759274765] | |2.0 |(11,[0,1,2,5,6,8,9],[1.0,4.0,1.0,4.0,9.0,1.0,2.0]) |[0.00419974114073206,0.9620399748900924,0.004199830998962131,0.0042814231963878655,0.004199801535688566,0.004199819689459903,0.004199830433436027,0.0041997822111186295,0.004199798534630995,0.0042799973694913] | |3.0 |(11,[0,1,3,6,8,9,10],[2.0,1.0,3.0,5.0,2.0,3.0,9.0]) |[0.0037148958393689426,0.5313564622081751,0.00371492700514763,0.4388535874884561,0.0037150382511682853,0.0037149506801198505,0.0037149808253623792,0.0037148901801274804,0.0037149076678115434,0.003785359854262734] | |4.0 |(11,[0,1,2,3,4,6,9,10],[3.0,1.0,1.0,9.0,3.0,2.0,1.0,3.0]) |[0.0040247360335797875,0.004348642552867576,0.004024775025300721,0.9633765038034603,0.004024773228145383,0.004024740478088116,0.00402477627651187,0.004024779618260475,0.004024784270292531,0.004101488713493013] | |5.0 |(11,[0,1,3,4,5,6,7,8,9],[4.0,2.0,3.0,4.0,5.0,1.0,1.0,1.0,4.0]) |[0.003714916663186164,0.004014116840889892,0.0037150323955768686,0.003787652360887051,0.0037149873236278505,0.003714958841217428,0.0037149705182189397,0.003715010255807931,0.0037149614099447853,0.9661933933906431] | |6.0 |(11,[0,1,3,6,8,9,10],[2.0,1.0,3.0,5.0,2.0,2.0,9.0]) |[0.003863635977009055,0.46449322935025966,0.0038636657354113126,0.5045241029221541,0.00386374420636613,0.0038636976398721237,0.003863727255143564,0.0038636207140121358,0.003863650494529744,0.003936925705242072] | |7.0 |(11,[0,1,2,3,4,5,6,9,10],[1.0,1.0,1.0,9.0,2.0,1.0,2.0,1.0,3.0])|[0.004390966123798511,0.004744425233669778,0.004391025010757086,0.9600440191238313,0.004391023986304413,0.00439098335688734,0.004391015731875719,0.004391018535344605,0.0043910130377361935,0.004474509859794904] | |8.0 |(11,[0,1,3,4,5,6,7],[4.0,4.0,3.0,4.0,2.0,1.0,3.0]) |[0.004391082402111978,0.0047448016288253025,0.004391206864616806,0.004477234571510909,0.004391077028823487,0.004391110359190354,0.004391102894332411,0.004391148031605367,0.004391148275359693,0.9600400879436237] | |9.0 |(11,[0,1,2,4,6,8,9,10],[2.0,8.0,2.0,3.0,2.0,2.0,7.0,2.0]) |[0.0033302167331450425,0.9698997342829896,0.003330238365882342,0.003394964707825143,0.0033302157712121493,0.0033302303649837654,0.0033302236683277224,0.0033302294595984666,0.0033302405714942906,0.0033937060745413443]| |10.0 |(11,[0,1,2,3,5,6,9,10],[1.0,1.0,1.0,9.0,2.0,2.0,3.0,3.0]) |[0.004199896541927494,0.00453848296824474,0.004200002237282065,0.9617819044818944,0.004200011124996577,0.004199942048495426,0.004199991764268097,0.004200001048497312,0.004199935367663148,0.004279832416731015] | |11.0 |(11,[0,1,4,5,6,7,9],[4.0,1.0,4.0,5.0,1.0,3.0,1.0]) |[0.004830560338779577,0.005219247495550288,0.004830593014957423,0.004924448157616727,0.00483055816775155,0.004830577856153918,0.004830584648561171,0.00483060040145597,0.004830612377397914,0.9560422175417754] | +-----+---------------------------------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
PCA降维
from __future__ import print_function from pyspark.ml.feature import PCA from pyspark.ml.linalg import Vectors from pyspark.sql import SparkSession
spark = SparkSession\
.builder\
.appName("PCAExample")\
.getOrCreate()
# 构建一份fake data
data = [(Vectors.sparse(5, [(1, 1.0), (3, 7.0)]),),#稀疏向量
(Vectors.dense([2.0, 0.0, 3.0, 4.0, 5.0]),),#稠密向量
(Vectors.dense([4.0, 0.0, 0.0, 6.0, 7.0]),)]
df = spark.createDataFrame(data, ["features"])
# PCA降维
pca = PCA(k=3, inputCol="features", outputCol="pcaFeatures")
model = pca.fit(df)
result = model.transform(df).select("pcaFeatures")
result.show(truncate=False)
spark.stop()
结果：
+-----------------------------------------------------------+ |pcaFeatures | +-----------------------------------------------------------+ |[1.6485728230883807,-4.013282700516296,-5.524543751369388] | |[-4.645104331781534,-1.1167972663619026,-5.524543751369387]| |[-6.428880535676489,-5.337951427775355,-5.524543751369389] | +-----------------------------------------------------------+
word2vec词嵌入
from __future__ import print_function from pyspark.ml.feature import Word2Vec from pyspark.sql import SparkSession
spark = SparkSession\
.builder\
.appName("Word2VecExample")\
.getOrCreate()
# 输入是bag of words形式
documentDF = spark.createDataFrame([
("Hi I heard about Spark".split(" "), ),
("I wish Java could use case classes".split(" "), ),
("Logistic regression models are neat".split(" "), )
], ["text"])
# 设置窗口长度等参数，词嵌入学习
word2Vec = Word2Vec(vectorSize=3, minCount=0, inputCol="text", outputCol="result")
model = word2Vec.fit(documentDF)
# 输出词和词向量
model.getVectors().show()
result = model.transform(documentDF)
for row in result.collect():
text, vector = row
print("Text: [%s] => \nVector: %s\n" % (", ".join(text), str(vector)))
spark.stop()
结果：
+----------+--------------------+ | word| vector| +----------+--------------------+ | heard|[0.08829052001237...| | are|[-0.1314301639795...| | neat|[0.09875790774822...| | classes|[-0.0047773420810...| | I|[0.15081347525119...| |regression|[-0.0732467696070...| | Logistic|[0.04169865325093...| | Spark|[-0.0096837198361...| | could|[-0.0907106027007...| | use|[-0.1245830804109...| | Hi|[0.03222155943512...| | models|[0.15642452239990...| | case|[-0.1072710305452...| | about|[0.13248910009860...| | Java|[0.08521263301372...| | wish|[0.02581630274653...| +----------+--------------------+ Text: [Hi, I, heard, about, Spark] => Vector: [0.0788261869922,-0.00265940129757,0.0531761907041] Text: [I, wish, Java, could, use, case, classes] => Vector: [-0.00935709210379,-0.015802019309,0.0161747672329] Text: [Logistic, regression, models, are, neat] => Vector: [0.0184408299625,-0.012609430775,0.0135096866637]
