1 package big.data.analyse.ml.randomforest 2 3 import org.apache.spark.ml.Pipeline 4 import org.apache.spark.ml.classification.{RandomForestClassificationModel, RandomForestClassifier} 5 import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator 6 import org.apache.spark.ml.feature.{IndexToString, VectorIndexer, StringIndexer} 7 import org.apache.spark.sql.SparkSession 8 9 /** 10 * 随机森林 11 * Created by zhen on 2018/9/20. 12 */ 13 object RandomForest { 14 def main(args: Array[String]) { 15 //创建spark对象 16 val spark = SparkSession.builder() 17 .appName("RandomForest") 18 .master("local[2]") 19 .getOrCreate() 20 //获取数据 21 val data = spark.read.format("libsvm") 22 .load("src/big/data/analyse/ml/randomforest/randomforest.txt") 23 //标识整个数据集的标识列和索引列 24 val labelIndexer = new StringIndexer() 25 .setInputCol("label") 26 .setOutputCol("indexedLabel") 27 .fit(data) 28 //设置树的最大层次 29 val featureIndexer = new VectorIndexer() 30 .setInputCol("features") 31 .setOutputCol("indexedFeatures") 32 .setMaxCategories(4) 33 .fit(data) 34 //拆分数据为训练集和测试集（7:3） 35 val Array(trainingData, testData) = data.randomSplit(Array(0.7, 0.3)) 36 testData.show(5) 37 //创建模型 38 val randomForest = new RandomForestClassifier() 39 .setLabelCol("indexedLabel") 40 .setFeaturesCol("indexedFeatures") 41 .setNumTrees(10) 42 //转化初始数据 43 val labelConverter = new IndexToString() 44 .setInputCol("prediction") 45 .setOutputCol("predictedLabel") 46 .setLabels(labelIndexer.labels) 47 //使用管道运行转换器和随机森林算法 48 val pipeline = new Pipeline() 49 .setStages(Array(labelIndexer, featureIndexer, randomForest, labelConverter)) 50 //训练模型 51 val model = pipeline.fit(trainingData) 52 //预测 53 val predictions = model.transform(testData) 54 //输出预测结果 55 predictions.select("predictedLabel", "label", "features").show(5) 56 //创建评估函数，计算错误率 57 val evaluator = new MulticlassClassificationEvaluator() 58 .setLabelCol("indexedLabel") 59 .setPredictionCol("prediction") 60 .setMetricName("accuracy") 61 62 val accuracy = evaluator.evaluate(predictions) 63 println("test error = " + (1.0 - accuracy)) 64 65 val rfModel = model.stages(2).asInstanceOf[RandomForestClassificationModel] 66 println("learned classification forest model: " + rfModel.toDebugString) 67 68 spark.stop() 69 } 70 }
使用数据：
0 128:51 129:159 130:253 131:159 132:50 155:48 156:238 157:252 158:252 1 159:124 160:253 161:255 162:63 186:96 187:244 188:251 189:253 190:62 1 125:145 126:255 127:211 128:31 152:32 153:237 154:253 155:252 156:71 1 153:5 154:63 155:197 181:20 182:254 183:230 184:24 209:20 210:254 1 152:1 153:168 154:242 155:28 180:10 181:228 182:254 183:100 209:190 0 130:64 131:253 132:255 133:63 157:96 158:205 159:251 160:253 161:205 1 159:121 160:254 161:136 186:13 187:230 188:253 189:248 190:99 213:4 1 100:166 101:222 102:55 128:197 129:254 130:218 131:5 155:29 156:249 0 155:53 156:255 157:253 158:253 159:253 160:124 183:180 184:253 185:25 0 128:73 129:253 130:227 131:73 132:21 156:73 157:251 158:251 159:251 1 155:178 156:255 157:105 182:6 183:188 184:253 185:216 186:14 210:14 0 154:46 155:105 156:254 157:254 158:254 159:254 160:255 161:239 162:41 0 152:56 153:105 154:220 155:254 156:63 178:18 179:166 180:233 181:253 1 130:7 131:176 132:254 133:224 158:51 159:253 160:253 161:223 185:4 0 155:21 156:176 157:253 158:253 159:124 182:105 183:176 184:251 185:25 1 151:68 152:45 153:131 154:131 155:131 156:101 157:68 158:92 159:44 0 125:29 126:170 127:255 128:255 129:141 151:29 152:198 153:255 154:255 0 153:203 154:254 155:252 156:252 157:252 158:214 159:51 160:20 180:62 1 98:64 99:191 100:70 125:68 126:243 127:253 128:249 129:63 152:30 1 125:26 126:240 127:72 153:25 154:238 155:208 182:209 183:226 184:14 0 155:62 156:91 157:213 158:255 159:228 160:91 161:12 182:70 183:230 1 157:42 158:228 159:253 160:253 185:144 186:251 187:251 188:251 212:89 1 128:62 129:254 130:213 156:102 157:253 158:252 159:102 160:20 184:102 0 154:28 155:195 156:254 157:254 158:254 159:254 160:254 161:255 162:61 0 123:8 124:76 125:202 126:254 127:255 128:163 129:37 130:2 150:13 0 127:68 128:254 129:255 130:254 131:107 153:11 154:176 155:230 156:253 1 157:85 158:255 159:103 160:1 185:205 186:253 187:253 188:30 213:205 1 126:94 127:132 154:250 155:250 156:4 182:250 183:254 184:95 210:250 1 124:32 125:253 126:31 152:32 153:251 154:149 180:32 181:251 182:188 1 129:39 130:254 131:255 132:254 133:140 157:136 158:253 159:253 160:22 0 123:59 124:55 149:71 150:192 151:254 152:250 153:147 154:17 176:123 1 128:58 129:139 156:247 157:247 158:25 183:121 184:253 185:156 186:3 1 129:28 130:247 131:255 132:165 156:47 157:221 158:252 159:252 160:164 0 156:13 157:6 181:10 182:77 183:145 184:253 185:190 186:67 207:11 0 127:28 128:164 129:254 130:233 131:148 132:11 154:3 155:164 156:254 0 129:105 130:255 131:219 132:67 133:67 134:52 156:20 157:181 158:253 0 125:22 126:183 127:252 128:254 129:252 130:252 131:252 132:76 151:85 1 155:114 156:206 157:25 183:238 184:252 185:55 211:222 212:252 213:55 1 127:73 128:253 129:253 130:63 155:115 156:252 157:252 158:144 183:217 1 120:85 121:253 122:132 123:9 147:82 148:241 149:251 150:251 151:128 1 126:15 127:200 128:255 129:90 154:42 155:254 156:254 157:173 182:42 0 182:32 183:57 184:57 185:57 186:57 187:57 188:57 189:57 208:67 209:18 0 127:42 128:235 129:255 130:84 153:15 154:132 155:208 156:253 157:253 1 156:202 157:253 158:69 184:253 185:252 186:121 212:253 213:252 214:69 1 156:73 157:253 158:253 159:253 160:124 184:73 185:251 186:251 187:251 1 124:111 125:255 126:48 152:162 153:253 154:237 155:63 180:206 181:253 0 99:70 100:255 101:165 102:114 127:122 128:253 129:253 130:253 131:120 1 124:29 125:197 126:255 127:84 152:85 153:251 154:253 155:83 180:86 1 159:31 160:210 161:253 162:163 187:198 188:252 189:252 190:162 213:10 1 131:159 132:255 133:122 158:167 159:228 160:253 161:121 185:64 186:23 0 153:92 154:191 155:178 156:253 157:242 158:141 159:104 160:29 180:26 1 128:53 129:250 130:255 131:25 156:167 157:253 158:253 159:25 182:3 0 122:63 123:176 124:253 125:253 126:159 127:113 128:63 150:140 151:253 0 153:12 154:136 155:254 156:255 157:195 158:115 159:3 180:6 181:175 1 128:255 129:253 130:57 156:253 157:251 158:225 159:56 183:169 184:254 0 151:23 152:167 153:208 154:254 155:255 156:129 157:19 179:151 180:253 1 130:24 131:150 132:233 133:38 156:14 157:89 158:253 159:254 160:254 0 125:120 126:253 127:253 128:63 151:38 152:131 153:246 154:252 155:252 1 127:155 128:253 129:126 155:253 156:251 157:141 158:4 183:253 184:251 0 101:88 102:127 103:5 126:19 127:58 128:20 129:14 130:217 131:19 152:7 0 127:37 128:141 129:156 130:156 131:194 132:194 133:47 153:11 154:132 0 154:32 155:134 156:218 157:254 158:254 159:254 160:217 161:84 176:44 1 124:102 125:252 126:252 127:41 152:102 153:250 154:250 155:202 180:10 0 124:20 125:121 126:197 127:253 128:64 151:23 152:200 153:252 154:252 1 127:20 128:254 129:255 130:37 155:19 156:253 157:253 158:134 183:19 0 235:40 236:37 238:7 239:77 240:137 241:136 242:136 243:136 244:136 1 128:166 129:255 130:187 131:6 156:165 157:253 158:253 159:13 183:15 1 128:117 129:128 155:2 156:199 157:127 183:81 184:254 185:87 211:116 1 129:159 130:142 156:11 157:220 158:141 184:78 185:254 186:141 212:111 0 124:66 125:254 126:254 127:58 128:60 129:59 130:59 131:50 151:73 1 129:101 130:222 131:84 157:225 158:252 159:84 184:89 185:246 186:208 0 124:41 125:254 126:254 127:157 128:34 129:34 130:218 131:255 132:206 0 96:56 97:247 98:121 124:24 125:242 126:245 127:122 153:231 154:253 0 125:19 126:164 127:253 128:255 129:253 130:118 131:59 132:36 153:78 1 129:232 130:255 131:107 156:58 157:244 158:253 159:106 184:95 185:253 1 127:63 128:128 129:2 155:63 156:254 157:123 183:63 184:254 185:179 1 130:131 131:255 132:184 133:15 157:99 158:247 159:253 160:182 161:15 0 125:57 126:255 127:253 128:198 129:85 153:168 154:253 155:251 156:253 0 127:12 128:105 129:224 130:255 131:247 132:22 155:131 156:254 157:254 1 130:226 131:247 132:55 157:99 158:248 159:254 160:230 161:30 185:125 1 130:166 131:253 132:124 133:53 158:140 159:251 160:251 161:180 185:12 1 129:17 130:206 131:229 132:44 157:2 158:125 159:254 160:123 185:95 1 130:218 131:253 132:124 157:84 158:236 159:251 160:251 184:63 185:236 1 124:102 125:180 126:1 152:140 153:254 154:130 180:140 181:254 182:204 0 128:87 129:208 130:249 155:27 156:212 157:254 158:195 182:118 183:225 1 126:134 127:230 154:133 155:231 156:10 182:133 183:253 184:96 210:133 1 125:29 126:85 127:255 128:139 153:197 154:251 155:253 156:251 181:254 1 125:149 126:255 127:254 128:58 153:215 154:253 155:183 156:2 180:41 1 130:79 131:203 132:141 157:51 158:240 159:240 160:140 185:88 186:252 1 126:94 127:254 128:75 154:166 155:253 156:231 182:208 183:253 184:147 0 127:46 128:105 129:254 130:254 131:224 132:59 133:59 134:9 155:196 1 125:42 126:232 127:254 128:58 153:86 154:253 155:253 156:58 181:86 1 156:60 157:229 158:38 184:187 185:254 186:78 211:121 212:252 213:254 1 101:11 102:150 103:72 129:37 130:251 131:71 157:63 158:251 159:71 0 127:45 128:254 129:254 130:254 131:148 132:24 133:9 154:43 155:254 0 125:218 126:253 127:253 128:255 129:149 130:62 151:42 152:144 153:236 0 127:60 128:96 129:96 130:48 153:16 154:171 155:228 156:253 157:251 0 126:32 127:202 128:255 129:253 130:253 131:175 132:21 152:84 153:144 1 130:218 131:170 132:108 157:32 158:227 159:252 160:232 185:129 186:25 1 130:116 131:255 132:123 157:29 158:213 159:253 160:122 185:189 186:25
结果（测试集&预测集）：
内部决策树结构：
test error = 0.34375 learned classification forest model: RandomForestClassificationModel (uid=rfc_0487ba2e1907) with 10 trees Tree 0 (weight 1.0): If (feature 185 <= 0.0) If (feature 157 <= 253.0) If (feature 149 <= 0.0) If (feature 210 in {3.0}) Predict: 0.0 Else (feature 210 not in {3.0}) If (feature 208 in {2.0}) Predict: 0.0 Else (feature 208 not in {2.0}) Predict: 0.0 Else (feature 149 > 0.0) Predict: 1.0 Else (feature 157 > 253.0) Predict: 1.0 Else (feature 185 > 0.0) If (feature 160 <= 0.0) If (feature 180 <= 0.0) Predict: 0.0 Else (feature 180 > 0.0) Predict: 1.0 Else (feature 160 > 0.0) Predict: 0.0 Tree 1 (weight 1.0): If (feature 156 <= 253.0) If (feature 187 <= 0.0) If (feature 133 in {2.0}) Predict: 1.0 Else (feature 133 not in {2.0}) If (feature 100 <= 11.0) If (feature 128 <= 139.0) Predict: 0.0 Else (feature 128 > 139.0) Predict: 1.0 Else (feature 100 > 11.0) Predict: 1.0 Else (feature 187 > 0.0) Predict: 0.0 Else (feature 156 > 253.0) Predict: 1.0 Tree 2 (weight 1.0): If (feature 158 <= 51.0) If (feature 182 <= 0.0) If (feature 127 <= 58.0) If (feature 129 <= 142.0) If (feature 154 <= 253.0) Predict: 0.0 Else (feature 154 > 253.0) Predict: 1.0 Else (feature 129 > 142.0) Predict: 1.0 Else (feature 127 > 58.0) Predict: 1.0 Else (feature 182 > 0.0) Predict: 0.0 Else (feature 158 > 51.0) If (feature 127 <= 62.0) Predict: 0.0 Else (feature 127 > 62.0) Predict: 1.0 Tree 3 (weight 1.0): If (feature 100 <= 11.0) If (feature 127 <= 0.0) If (feature 151 <= 162.0) If (feature 159 <= 0.0) If (feature 125 <= 48.0) Predict: 0.0 Else (feature 125 > 48.0) Predict: 0.0 Else (feature 159 > 0.0) Predict: 0.0 Else (feature 151 > 162.0) Predict: 1.0 Else (feature 127 > 0.0) If (feature 131 <= 0.0) If (feature 153 <= 42.0) Predict: 0.0 Else (feature 153 > 42.0) If (feature 154 <= 228.0) Predict: 1.0 Else (feature 154 > 228.0) Predict: 0.0 Else (feature 131 > 0.0) Predict: 1.0 Else (feature 100 > 11.0) Predict: 1.0 Tree 4 (weight 1.0): If (feature 152 <= 0.0) If (feature 158 <= 0.0) If (feature 151 <= 0.0) Predict: 1.0 Else (feature 151 > 0.0) Predict: 0.0 Else (feature 158 > 0.0) If (feature 182 <= 15.0) If (feature 153 <= 0.0) Predict: 0.0 Else (feature 153 > 0.0) Predict: 1.0 Else (feature 182 > 15.0) Predict: 1.0 Else (feature 152 > 0.0) If (feature 124 <= 0.0) Predict: 1.0 Else (feature 124 > 0.0) If (feature 123 <= 24.0) If (feature 125 <= 232.0) Predict: 0.0 Else (feature 125 > 232.0) Predict: 1.0 Else (feature 123 > 24.0) Predict: 0.0 Tree 5 (weight 1.0): If (feature 157 <= 0.0) If (feature 101 <= 0.0) If (feature 129 <= 0.0) If (feature 183 <= 0.0) If (feature 152 <= 231.0) Predict: 0.0 Else (feature 152 > 231.0) Predict: 0.0 Else (feature 183 > 0.0) Predict: 0.0 Else (feature 129 > 0.0) Predict: 1.0 Else (feature 101 > 0.0) Predict: 1.0 Else (feature 157 > 0.0) If (feature 155 <= 165.0) Predict: 0.0 Else (feature 155 > 165.0) Predict: 1.0 Tree 6 (weight 1.0): If (feature 153 <= 253.0) If (feature 125 <= 240.0) If (feature 158 <= 3.0) If (feature 182 <= 0.0) If (feature 179 <= 6.0) Predict: 1.0 Else (feature 179 > 6.0) Predict: 0.0 Else (feature 182 > 0.0) If (feature 128 <= 139.0) Predict: 0.0 Else (feature 128 > 139.0) Predict: 0.0 Else (feature 158 > 3.0) If (feature 155 <= 58.0) Predict: 0.0 Else (feature 155 > 58.0) If (feature 175 in {1.0}) Predict: 1.0 Else (feature 175 not in {1.0}) Predict: 0.0 Else (feature 125 > 240.0) If (feature 129 <= 0.0) If (feature 154 <= 0.0) Predict: 1.0 Else (feature 154 > 0.0) Predict: 0.0 Else (feature 129 > 0.0) Predict: 1.0 Else (feature 153 > 253.0) Predict: 1.0 Tree 7 (weight 1.0): If (feature 131 <= 67.0) If (feature 155 <= 102.0) If (feature 129 <= 226.0) If (feature 129 <= 62.0) If (feature 127 <= 58.0) Predict: 0.0 Else (feature 127 > 58.0) Predict: 1.0 Else (feature 129 > 62.0) Predict: 0.0 Else (feature 129 > 226.0) Predict: 1.0 Else (feature 155 > 102.0) If (feature 128 <= 224.0) If (feature 184 <= 25.0) If (feature 157 <= 0.0) Predict: 1.0 Else (feature 157 > 0.0) Predict: 1.0 Else (feature 184 > 25.0) Predict: 0.0 Else (feature 128 > 224.0) If (feature 131 <= 0.0) Predict: 0.0 Else (feature 131 > 0.0) Predict: 1.0 Else (feature 131 > 67.0) Predict: 0.0 Tree 8 (weight 1.0): If (feature 182 <= 180.0) If (feature 179 <= 62.0) If (feature 128 <= 101.0) If (feature 156 <= 225.0) If (feature 149 <= 0.0) Predict: 0.0 Else (feature 149 > 0.0) Predict: 1.0 Else (feature 156 > 225.0) If (feature 155 <= 202.0) Predict: 0.0 Else (feature 155 > 202.0) Predict: 1.0 Else (feature 128 > 101.0) If (feature 183 <= 0.0) If (feature 128 <= 254.0) Predict: 1.0 Else (feature 128 > 254.0) Predict: 0.0 Else (feature 183 > 0.0) Predict: 0.0 Else (feature 179 > 62.0) Predict: 0.0 Else (feature 182 > 180.0) If (feature 156 <= 105.0) Predict: 0.0 Else (feature 156 > 105.0) Predict: 1.0 Tree 9 (weight 1.0): If (feature 96 in {1.0}) Predict: 1.0 Else (feature 96 not in {1.0}) If (feature 185 <= 67.0) If (feature 160 <= 12.0) If (feature 178 in {1.0}) Predict: 1.0 Else (feature 178 not in {1.0}) If (feature 126 <= 0.0) Predict: 0.0 Else (feature 126 > 0.0) Predict: 1.0 Else (feature 160 > 12.0) If (feature 155 <= 0.0) Predict: 0.0 Else (feature 155 > 0.0) Predict: 1.0 Else (feature 185 > 67.0) Predict: 0.0
总结：可知该随机森林共有10棵树组成，预测结果为10棵树的投票为准。每棵树的最大层次为4，这是为了避免层次过高带来的计算压力和过拟合！