使用隨機森林 model 預測測試數據時出錯

Question

我正在嘗試使用隨機森林 model 對我的測試數據進行預測。 但是，我收到以下錯誤：

predict.randomForest（modelFit，newdata）中的錯誤：newdata 中的缺失值

我不確定怎么會有缺失值，因為我已經檢查過並且似乎沒有任何 NA。

我的代碼：

tr.Control <- trainControl(method = "repeatedcv",
                           number = 10,
                           repeats = 5,
                           )

rf3 <- train(
  Lifeexp ~ .,
  data = train.dat2,
  method = "rf",
  trControl = tr.Control ,
  preProc = c("center", "scale"),  
  ntree = 1500,
  tuneGrid = expand.grid(mtry = seq(1, ncol(train.dat2)-1)
                         )
)

rf.pred <- predict(rf3, newdata = test.dat2, na.action = na.pass, type = "raw")

RMSE.tree = RMSE(rf.pred, test.dat2$Lifeexp), 
  Rsquare.tree = R2(rf.pred, test.dat2$Lifeexp)

火車.dat2：

structure(list(Status = c("Low", "Low", "Low", "Low", "Low", 
"Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", 
"Low", "Low", "Low", "Low", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High"), GDP = c(402.1030419, 
442.2030419, 543.3030419, 520.8966027, 254.2432569, 124.4608003, 
341.5541149, 772.3135303, 478.6685897, 191.8789042, 592.4010975, 
1033.912431, 138.4288795, 622.4988457, 642.7767443, 317.3893069, 
269.8711377, 709.5819646, 585.07655, 780.190201, 3122.362815, 
3893.596078, 1166.610276, 1674.825261, 3690.113268, 4241.788782, 
2441.741991, 4043.662051, 9040.566251, 963.8417858, 2234.579866, 
10330.61561, 1944.137621, 2136.440243, 567.5286729, 567.930736, 
2292.445156, 2028.18197, 371.6785662, 519.5343268, 987.409723, 
1482.403063, 1196.586858, 1955.588006, 6941.235848, 1038.90854, 
3102.713363, 3139.966054, 3032.427138, 7328.615629, 869.6965166, 
2799.648876, 617.2304355, 1126.683318, 4094.362119, 7708.100996, 
10385.96443, 11683.94962, 718.1878292, 3243.231125, 3100.280468, 
11286.24302, 8920.762105, 201.4671636, 785.5022829, 1510.324871, 
1831.001912, 8141.913127, 12027.36588, 6967.24523, 7691.345097, 
3233.295943, 367.5566093, 1357.563719, 1489.876911, 977.2736357, 
1508.942737, 2007.736363, 5076.342992, 7273.563207, 948.3318545, 
2146.996385, 95.18825018, 390.0933261, 2566.59695, 52022.1256, 
57373.68668, 19095.467, 28149.87001, 39435.8399, 20600.37525, 
23041.53473, 44141.87814, 47518.63604, 24190.24962, 46232.98962, 
26891.44645, 61350.34791, 28364.64508, 50152.34014, 22303.96133, 
23635.92922, 41531.9342, 47603.02763, 9600.18513, 12042.95373, 
26917.75898, 20324.25356, 20087.59199, 36000.52012, 25423.07201, 
32018.06325, 43024.92384, 73191.11632, 12663.36453, 30693.59308, 
18440.37852, 38577.38166, 33994.40657, 21290.86038, 50950.03434, 
53024.05921, 13663.02162, 13641.10272, 41945.33167, 1731.209509, 
4492.727604, 11861.75616, 47236.96023, 23509.54339, 26123.97387, 
74605.77451), Health = c(22.23474948, 36.44474948, 45.58774948, 
46.38774948, 3.333203815, 5.359203815, 16.69390488, 19.46990488, 
33.22835541, 5.300580788, 29.97179604, 33.59179604, 5.971383095, 
62.66848373, 67.22848373, 8.23568, 14.98141193, 32.6487999, 10.22661548, 
16.19961548, 92.18703461, 98.65987461, 143.7665911, 159.7515106, 
308.6578979, 402.5568979, 99.5689502, 111.4155502, 292.8907166, 
198.2263198, 221.1403198, 705.336568, 176.6524443, 200.7054443, 
12.56211728, 17.72411728, 76.7208786, 98.4562786, 9.55682529, 
16.01162529, 26.5686245, 33.565445, 69.66563616, 89.45643616, 
275.2236792, 32.77552414, 122.5689168, 198.7124574, 221.7829742, 
539.567627, 43.70681763, 108.6149597, 33.2254878, 42.36598, 60.2569, 
705.1993408, 891.1377563, 992.5689563, 31.84200096, 77.2356478, 
277.45864, 891.7641602, 932.325129, 15.23564, 54.30473709, 74.231488, 
200.564125, 665.2514038, 755.36985, 384.9183044, 445.20158, 262.5267029, 
11.56898, 45.25077438, 109.0749969, 122.02145, 42.568412, 62.25963211, 
172.0576935, 200.562134, 91.17743683, 120.236549, 11.23587, 18.82835197, 
99.23568, 4952.777344, 5236.3654, 1101.36589, 1674.2854, 3309.480957, 
1654.5687, 1845.321045, 4449.542969, 5000.36545, 1998.634277, 
6054.23658, 1900.2356, 7025.36987, 1000.5689, 5036.2356, 1233.36545, 
2334.651855, 4597.244629, 5698.2547, 1500.3698, 2000.23564, 2573.740234, 
3002.36547, 1520.453613, 3214.546387, 1569.3254, 2873.848145, 
3644.802734, 4587.235478, 1122.02145, 2211.019043, 462.5890808, 
1061.365601, 1256.56897, 1987.2145, 5186.632813, 6547.2356, 990.32658, 
1053.891602, 4201.3698, 122.02145, 238.0044861, 712.2356, 1513.565918, 
2015.18042, 2985.23, 8021.80957), Govthealth = c(1.25689, 2.032658, 
2.495758057, 2.965478, 1.985478, 2.209019899, 2.882325411, 3.21458, 
7.3134408, 1.032568, 5.433434963, 7.235478, 1.239725351, 8.535984039, 
10.323589, 1.236589, 3.562868595, 4.673761368, 2.32547, 4.648055553, 
23.70949936, 33.235687, 51025478, 71.8605423, 205.9026794, 295.2356, 
31.2587, 51.99817276, 154.70401, 56.32588, 73.30036926, 399.23568, 
66.3265, 99.82849121, 2.23568, 3.246135235, 10.43734169, 15.235478, 
3.569877, 5.623521328, 5.849419594, 8.32665, 35.3654457, 44.96020508, 
195.3657, 14.55177689, 35.235698, 61.02356, 81.59127045, 284.7705994, 
23.43979454, 43.92045593, 22.36587, 30.42416763, 181.3415375, 
385.9675598, 576.0806274, 602.3258, 25.36730576, 66.235687, 92.2147, 
401.4833984, 502.3698, 2.0214578, 10.70767879, 15.36987, 112.3698, 
481.0765686, 502.36987, 226.7909851, 300.65478, 55.95266342, 
2.36547, 11.85855961, 35.50076675, 45.235698, 25.36954, 34.36005783, 
126.9312592, 156.3257, 23.53768349, 39.235687, 4.235687, 6.570708275, 
45.36987, 3399.406006, 4500.321547, 990.36547, 1368.160278, 2804.857178, 
1000.365, 1375.334717, 3458.573975, 4120.325, 1456.037842, 4100.368, 
1500.36578, 6925.325445, 990.58795, 4125.25658, 998.25998, 1827.566895, 
3482.541016, 4800.3256, 989.325, 1254.325, 1756.99939, 1998.23569, 
1104.429321, 2521.927002, 1800.3256, 2315.543701, 2931.431641, 
331.0256, 548.32, 1388.55896, 351.3133545, 898.4367065, 997.02145, 
956.32547, 3488.651855, 4400.23556, 558.36987, 785.0509033, 3000.3658, 
100.36987, 162.3498688, 162.365, 543.0645752, 1458.283813, 2000.3694, 
2495.23877), Privhealth = c(14.3698, 25.36698, 36.01279831, 49.36875, 
1.23569, 2.278559208, 8.061329842, 10.3658, 5.059076786, 3.25698, 
20.38587761, 30.65877, 4.726452827, 22.79703331, 32.65878, 6.32589, 
10.38636589, 19.33849907, 8.326589, 11.07592678, 67.27728271, 
74.23658, 63.235698, 83.74517059, 88.83229828, 96.32568, 49.32658, 
59.41738892, 138.1631165, 100.23564, 147.8399658, 300.23568, 
71.02584, 90.6206665, 8.365984, 11.47062778, 61.48280716, 74.254785, 
7.235647, 10.26313496, 19.40570831, 23.65879, 33.25478, 44.17641068, 
189.32658, 17.06592751, 75.325689, 89.32658, 136.7345276, 238.6507721, 
19.86775017, 63.43461227, 7.325478, 19.23568, 25.321547, 319.0157471, 
311.9694214, 442.03695, 3.889117956, 15.3654, 115.02365, 488.0875244, 
552.0325698, 10.3658, 36.04922485, 45.362154, 45.23548, 182.7733917, 
202.3654, 142.2067719, 202.325, 197.0276337, 9.32658, 32.95304871, 
70.28269196, 90.3256, 15.021457, 27.89465141, 44.9021492, 60.32568, 
43.03323364, 60.325845, 8.325698, 11.45799065, 60.32568, 1553.358765, 
2330.2354, 201.0214578, 305.5347595, 503.7982178, 301.23565, 
469.9864197, 990.9689331, 1200.36987, 542.5964966, 1823.021457, 
312.0215478, 1100.32145, 301.02145, 1100.3256, 320.365478, 507.0849609, 
1114.720093, 2001.23548, 401.14567, 662.03214, 816.2644653, 998.32546, 
416.0243225, 692.6192017, 402.32564, 558.3044434, 713.3709106, 
998.32658, 302.0214, 793.8995972, 111.2757187, 162.9289398, 212.3657, 
442.32598, 1698.060913, 2226.32568, 145.2365, 268.8859863, 902.32568, 
42.36587, 75.64861298, 332.65478, 970.5014648, 556.8964233, 700.32658, 
5526.447266), Lifeexp = c(50.331, 55.841, 61.028, 64.486, 47.099, 
51.941, 61.627, 66.24, 55.564, 54.404, 67.611, 70.478, 61.974, 
57.099, 62.973, 45.746, 48.069, 55.251, 62.32, 65.772, 69.205, 
71.509, 69.872, 71.73, 73.428, 74.405, 70.865, 72.594, 74.493, 
61.529, 70.173, 78.627, 61.608, 52.192, 45.9, 46.267, 50.896, 
54.332, 60.1, 62.82, 65.264, 67.114, 66.165, 71.111, 76.516, 
68.793, 71.095, 63.307, 56.048, 57.669, 71.333, 75.439, 71.836, 
73.955, 76.562, 73.576, 75.278, 76.52, 60.884, 71.46, 66.343, 
73.619, 75.672, 53.595, 66.56, 69.57, 75.654, 78.769, 80.095, 
74.619, 77.672, 71.46, 57.865, 66.693, 62.764, 65.095, 70.248, 
70.623, 74.184, 76.931, 50.64, 61.195, 70.551, 73.025, 75.317, 
81.69512195, 82.74878049, 75.8804878, 77.74146341, 80.40243902, 
76.05195122, 77.72195122, 80.18292683, 81.59512195, 79.13658537, 
81.94878049, 74.80536585, 81.35121951, 74.81317073, 81.83414634, 
75.2277561, 77.92682927, 79.98780488, 80.99268293, 76.93902439, 
77.88780488, 80.38780488, 81.28780488, 79.77804878, 82.03658537, 
78.03634146, 79.65365854, 81.89756098, 82.66097561, 76.60731707, 
81.60243902, 73.142, 74.358, 75.398, 76.87804878, 80.70243902, 
81.76097561, 75.37804878, 78.63658537, 81.85853659, 70.8902439, 
73.74878049, 75.29512195, 81.54146341, 79.42195122, 81.02926829, 
82.24634146)), row.names = c(1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 
11L, 13L, 15L, 16L, 18L, 23L, 24L, 25L, 26L, 27L, 29L, 30L, 31L, 
32L, 33L, 34L, 35L, 36L, 37L, 38L, 39L, 41L, 42L, 44L, 45L, 46L, 
49L, 50L, 51L, 52L, 53L, 54L, 55L, 56L, 57L, 58L, 60L, 62L, 64L, 
65L, 66L, 67L, 70L, 71L, 73L, 74L, 75L, 78L, 79L, 80L, 82L, 84L, 
85L, 87L, 88L, 89L, 91L, 92L, 93L, 95L, 96L, 99L, 100L, 103L, 
105L, 107L, 111L, 112L, 113L, 114L, 115L, 116L, 119L, 120L, 121L, 
122L, 124L, 127L, 128L, 129L, 130L, 131L, 133L, 134L, 135L, 136L, 
138L, 140L, 141L, 144L, 145L, 148L, 149L, 150L, 151L, 152L, 153L, 
154L, 155L, 156L, 158L, 159L, 161L, 162L, 163L, 164L, 165L, 167L, 
170L, 171L, 172L, 173L, 175L, 176L, 177L, 178L, 180L, 181L, 182L, 
185L, 187L, 191L, 192L, 195L), class = "data.frame")

測試.dat2：

structure(list(Status = c("Low", "Low", "Low", "Low", "Low", 
"Low", "Low", "Low", "Low", "Low", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "Middle", "Middle", 
"Middle", "Middle", "Middle", "Middle", "Middle", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High", "High", 
"High", "High", "High", "High", "High", "High", "High"), GDP = c(199.9863423, 
156.3857186, 389.3980332, 229.4902871, 497.6320261, 749.552711, 
826.6215305, 248.0293672, 261.8689977, 899.6599081, 11373.233, 
7076.662423, 5324.61704, 5931.453886, 5082.354757, 715.9137121, 
2124.05677, 6374.028196, 463.6186318, 4102.48135, 5268.848504, 
4333.482973, 564.7796095, 2258.183141, 3749.75325, 302.5771636, 
3772.870012, 2860.43156, 4787.780171, 1614.640122, 749.9085236, 
4717.143026, 443.3141934, 2009.978857, 483.952592, 366.1728076, 
841.9729898, 563.0577411, 1317.890706, 18211.27459, 21679.24784, 
42943.90227, 21448.36196, 47450.31847, 30743.54768, 58041.39844, 
24285.46682, 46459.97325, 20825.78421, 34483.204, 21043.57493, 
41715.02928, 8794.631229, 26149.41108, 33692.01083, 12599.53358, 
15420.91116, 23852.32703, 64581.94402, 9107.477079, 10201.30354, 
38428.3855, 37868.296, 82796.54716), Health = c(6.22435541, 8.909747124, 
39.22274712, 8.625580788, 4.22284155, 42.34384155, 47.44484155, 
10.74555809, 18.80055809, 45.32365, 324.6654166, 602.659668, 
504.5536499, 594.8854499, 239.3392792, 22.55662414, 91.84031677, 
624.335527, 30.56891763, 128.3355597, 74.23569, 505.4589408, 
22.23569, 69.80043793, 311.6526794, 19.73552704, 251.0935822, 
211.589745, 250.7455292, 35.25698, 47.90106964, 292.54782, 18.56432343, 
70.5685123, 10.56888, 17.38329887, 50.66987, 75.201547, 78.18682861, 
1022.5487, 1632.427612, 4002.325, 1452.369, 5044.135254, 2496.047119, 
6011.536621, 1655.866211, 4099.587891, 1125.365, 4400.325, 1496.87854, 
3000.23568, 336.2356, 2023.143677, 3216.223633, 809.1994019, 
956.21547, 820.6981812, 1989.235, 446.3265, 796.6470337, 2985.12, 
3737.802979, 9658.23), Govthealth = c(2.65987, 3.350677967, 8.32365, 
1.337858081, 0.235689, 8.714180946, 11.02365, 2.356894, 4.656533241, 
5.958777, 198.23568, 319.1759033, 207.0215302, 302.654789, 123.2336197, 
9.32658, 29.2992878, 300.5689, 12.02589, 52.658912, 22.03256, 
222.325689, 16.3258, 50.29269791, 129.758316, 3.900079966, 163.0175018, 
102.369, 156.8104706, 4.36987, 5.465222836, 75.36987, 3.839128733, 
14.32589, 3.25478, 5.880064487, 12.36547, 18.02584, 30.97570801, 
990.365478, 1116.231445, 3201.0245, 996.598723, 3721.796387, 
2074.39917, 5042.459961, 1229.708252, 3167.418213, 889.32658, 
3698.23598, 944.5585938, 1998.02365, 200.365778, 1396.733398, 
2517.370117, 577.3640747, 662.32589, 298.1834717, 702.369, 456.325, 
568.7339478, 889.36547, 1045.900513, 3987.3654), Privhealth = c(1.36589, 
1.832908154, 7.325698, 5.431494236, 2.36589, 29.85413742, 35.3698, 
4.23568, 8.9836483, 22.3658, 152.36589, 263.3545532, 225.5363922, 
301.325478, 111.575592, 10.23568, 60.89479446, 336.02145, 12.36587, 
75.36987, 34.3265, 223.02145, 2.0215478, 11.81901455, 180.9026947, 
15.41190529, 85.28456879, 45.321478, 86.49634552, 25.36987, 39.00668716, 
220.32145, 14.22738075, 49.326545, 7.02145, 11.50323391, 20.36587, 
33.021456, 45.45627975, 400.23568, 516.1798096, NA, 400.32547, 
1322.338745, 421.6481018, 969.076416, 426.0691833, 931.8737793, 
302.1245, 886.02154, 517.4750366, 889.32547, 90.3256, 626.4102173, 
698.8658447, 231.8352966, 301.0324, 522.5147705, 1236.021458, 
117.3658, 227.9130707, 1965.3256, 2691.985107, 6600.3256), Lifeexp = c(46.096, 
45.09, 63.798, 62.288, 58.824, 68.736, 70.879, 45.853, 46.229, 
58.893, 75.997, 75.905, 56.665, 63.373, 74.41, 66.366, 69.823, 
63.857, 69.509, 76.812, 78.458, 71.594, 52.878, 68.384, 70.116, 
58.432, 77.452, 66.843, 71.116, 70.386, 69.902, 73.6, 62.505, 
69.416, 55.5, 58.472, 58.1, 44.649, 74.837, 76.99463415, 79.23414634, 
81.35609756, 77.42195122, 81.24634146, 76.59268293, 79.1, 77.46585366, 
79.87073171, 76.97073171, 82.94634146, 78.95365854, 82.80243902, 
72.15, 77.98780488, 80.70243902, 76.24634146, 77.75365854, 77.95121951, 
83.14634146, 73.20487805, 75.41219512, 77.24243902, 79.6804878, 
83.55121951)), row.names = c(9L, 10L, 12L, 14L, 17L, 19L, 20L, 
21L, 22L, 28L, 40L, 43L, 47L, 48L, 59L, 61L, 63L, 68L, 69L, 72L, 
76L, 77L, 81L, 83L, 86L, 90L, 94L, 97L, 98L, 101L, 102L, 104L, 
106L, 108L, 109L, 110L, 117L, 118L, 123L, 125L, 126L, 132L, 137L, 
139L, 142L, 143L, 146L, 147L, 157L, 160L, 166L, 168L, 169L, 174L, 
179L, 183L, 184L, 186L, 188L, 189L, 190L, 193L, 194L, 196L), class = "data.frame")

Answer 1

您可以先刪除 NA 然后使用predict 、 RMSE 、 R2 function 之類的

library(caret)

tr.Control <- trainControl(method = "repeatedcv",
                           number = 10,
                           repeats = 5)

rf3 <- caret::train(Lifeexp~.,
  data = train.dat2,
  method = "rf",
  trControl = tr.Control ,
  preProcess = c("center", "scale"),  
  ntree = 1500,
  tuneGrid = expand.grid(mtry = seq(1, ncol(train.dat2)-1))
)

#Remove the NA from the data freme
test.dat2 <- na.omit(test.dat2)
rf.pred <- predict(rf3, newdata = test.dat2, type = "raw")

RMSE.tree = RMSE(rf.pred, test.dat2$Lifeexp)
Rsquare.tree = R2(rf.pred, test.dat2$Lifeexp)