176 KiB
176 KiB
In [2]:
import pandas as pd import lightgbm as lgb import numpy as np from sklearn.model_selection import train_test_split from sklearn.metrics import mean_absolute_error, mean_squared_error, mean_absolute_percentage_error, r2_score import seaborn as sns import matplotlib.pyplot as plt #新增加的两行 from pylab import mpl # 设置显示中文字体 mpl.rcParams["font.sans-serif"] = ["SimHei"] mpl.rcParams["axes.unicode_minus"] = False
In [3]:
total_data = pd.read_csv('./data/train_data.csv') total_data.head()
Out[3]:
| days | 发电量（千瓦时） | 供热量（吉焦） | 机组运行时间（小时） | 硫分（%） | 脱硫剂使用量（吨） | 脱硫设施运行时间（小时） | 脱硝还原剂消耗量（吨） | 脱硝运行时间（小时） | 燃料消耗量（吨） | ... | cSO2 | cO2 | csmoke | flow | rNOx | rO2 | temp | rSO2 | rsmoke | day_of_year | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 2018-10-01 | 156796.0 | 6536.83 | 24.0 | 0.51 | 5.06 | 24.0 | 2.98 | 24.0 | 323 | ... | 2.148473e+07 | 3.745944e+07 | 6.519466e+05 | 162345.192917 | 28.981417 | 9.900000 | 51.250000 | 5.581667 | 0.209167 | 273 |
| 1 | 2018-10-02 | 133984.0 | 2484.64 | 24.0 | 0.51 | 5.04 | 24.0 | 2.97 | 24.0 | 218 | ... | 1.587722e+07 | 2.832146e+07 | 3.656575e+05 | 140175.330833 | 22.220750 | 9.400000 | 50.679167 | 4.364167 | 0.190417 | 274 |
| 2 | 2018-10-03 | 134023.0 | 3020.83 | 24.0 | 0.51 | 5.04 | 24.0 | 2.95 | 24.0 | 212 | ... | 2.829086e+07 | 3.174159e+07 | 5.181773e+05 | 154686.184167 | 24.816708 | 8.550000 | 52.808333 | 7.580000 | 0.139583 | 275 |
| 3 | 2018-10-04 | 124765.0 | 5599.23 | 24.0 | 0.51 | 5.03 | 24.0 | 2.98 | 24.0 | 223 | ... | 1.030569e+07 | 2.511504e+07 | 2.299438e+06 | 120345.545833 | 21.875125 | 10.202083 | 48.854167 | 2.808958 | 0.893333 | 276 |
| 4 | 2018-10-05 | 134414.0 | 4702.65 | 24.0 | 0.51 | 5.06 | 24.0 | 3.01 | 24.0 | 243 | ... | 1.830254e+06 | 4.106346e+07 | 6.230433e+06 | 162533.103542 | 25.605917 | 11.497917 | 45.783333 | 0.393333 | 2.141875 | 277 |
5 rows × 165 columns
In [4]:
# 先去掉一些异常值 use_data = total_data[(total_data['机组运行时间（小时）'] == 24) & (total_data['脱硝运行时间（小时）'] == 24)].set_index('days').drop( columns=['机组运行时间（小时）', '脱硫设施运行时间（小时）', '脱硝运行时间（小时）']) use_data['week_of_year'] = use_data.day_of_year / 7 use_data.drop(columns=['day_of_year'], inplace=True)
In [5]:
r_cols = [x for x in use_data.columns if x.startswith('r')] r_cols
Out[5]:
['rNOx', 'rO2', 'rSO2', 'rsmoke']
In [6]:
for col in use_data.columns: if use_data[col].max() > 10: use_data[col] = np.log1p(use_data[col]) if col == '燃料消耗量（吨）': use_data[col] = np.log1p(use_data[col])
In [7]:
feature_cols = [x for x in use_data.columns if x != '燃料消耗量（吨）'] feature_cols = [x for x in feature_cols if '（' not in x] feature_cols = use_data.columns[-11:] feature_cols
Out[7]:
Index(['cNOx', 'cSO2', 'cO2', 'csmoke', 'flow', 'rNOx', 'rO2', 'temp', 'rSO2',
'rsmoke', 'week_of_year'],
dtype='object')
In [8]:
train_data, valid_data = train_test_split(use_data, test_size=0.15, shuffle=True, random_state=666)
In [9]:
X_train, Y_train = train_data[feature_cols], train_data['燃料消耗量（吨）'] X_valid, Y_valid = valid_data[feature_cols], valid_data['燃料消耗量（吨）'] X_test, Y_test = valid_data[feature_cols], valid_data['燃料消耗量（吨）']
In [10]:
lgb_train = lgb.Dataset(X_train, Y_train) lgb_eval = lgb.Dataset(X_valid, Y_valid) lgb_test = lgb.Dataset(X_test, Y_test)
In [11]:
params = { 'task': 'train', 'boosting_type': 'gbdt', # 设置提升类型 'objective': 'regression_l2', # 目标函数 'metric': {'rmse'}, # 评估函数 'max_depth': 10, 'num_leaves': 31, # 叶子节点数 'learning_rate': 0.01, # 学习速率 'feature_fraction': 0.9, # 建树的特征选择比例 'bagging_fraction': 0.8, # 建树的样本采样比例 'bagging_freq': 5, # k 意味着每 k 次迭代执行bagging 'verbose': -1 # <0 显示致命的, =0 显示错误 (警告), >0 显示信息 }
In [12]:
gbm = lgb.train(params, lgb_train, num_boost_round=2000, valid_sets=lgb_eval, early_stopping_rounds=100)
[1] valid_0's rmse: 0.0361047 Training until validation scores don't improve for 100 rounds [2] valid_0's rmse: 0.0359398 [3] valid_0's rmse: 0.0358123 [4] valid_0's rmse: 0.0356709 [5] valid_0's rmse: 0.0355485 [6] valid_0's rmse: 0.035388 [7] valid_0's rmse: 0.0352781 [8] valid_0's rmse: 0.0351619 [9] valid_0's rmse: 0.035026 [10] valid_0's rmse: 0.0348736 [11] valid_0's rmse: 0.0347371 [12] valid_0's rmse: 0.0346186 [13] valid_0's rmse: 0.0345018 [14] valid_0's rmse: 0.0343709 [15] valid_0's rmse: 0.0342329 [16] valid_0's rmse: 0.0341227 [17] valid_0's rmse: 0.0340256 [18] valid_0's rmse: 0.033924 [19] valid_0's rmse: 0.0338187 [20] valid_0's rmse: 0.0337153 [21] valid_0's rmse: 0.033616 [22] valid_0's rmse: 0.0335067 [23] valid_0's rmse: 0.0334041 [24] valid_0's rmse: 0.0332995 [25] valid_0's rmse: 0.0331921 [26] valid_0's rmse: 0.0330884 [27] valid_0's rmse: 0.032978 [28] valid_0's rmse: 0.0328855 [29] valid_0's rmse: 0.032781 [30] valid_0's rmse: 0.032678 [31] valid_0's rmse: 0.0325855 [32] valid_0's rmse: 0.0325005 [33] valid_0's rmse: 0.0324143 [34] valid_0's rmse: 0.0323269 [35] valid_0's rmse: 0.0322377 [36] valid_0's rmse: 0.0321433 [37] valid_0's rmse: 0.0320294 [38] valid_0's rmse: 0.0319328 [39] valid_0's rmse: 0.031826 [40] valid_0's rmse: 0.0317349 [41] valid_0's rmse: 0.0316664 [42] valid_0's rmse: 0.0315913 [43] valid_0's rmse: 0.0315254 [44] valid_0's rmse: 0.031468 [45] valid_0's rmse: 0.0314055 [46] valid_0's rmse: 0.0313306 [47] valid_0's rmse: 0.0312563 [48] valid_0's rmse: 0.0311767 [49] valid_0's rmse: 0.0311059 [50] valid_0's rmse: 0.0310381 [51] valid_0's rmse: 0.0309327 [52] valid_0's rmse: 0.0308338 [53] valid_0's rmse: 0.0307449 [54] valid_0's rmse: 0.030644 [55] valid_0's rmse: 0.030561 [56] valid_0's rmse: 0.0304835 [57] valid_0's rmse: 0.0304175 [58] valid_0's rmse: 0.0303561 [59] valid_0's rmse: 0.0302768 [60] valid_0's rmse: 0.0302058 [61] valid_0's rmse: 0.0301515 [62] valid_0's rmse: 0.0301012 [63] valid_0's rmse: 0.0300401 [64] valid_0's rmse: 0.0299948 [65] valid_0's rmse: 0.0299354 [66] valid_0's rmse: 0.0298821 [67] valid_0's rmse: 0.0298 [68] valid_0's rmse: 0.0297534 [69] valid_0's rmse: 0.0297084 [70] valid_0's rmse: 0.0296541 [71] valid_0's rmse: 0.0295918 [72] valid_0's rmse: 0.0295303 [73] valid_0's rmse: 0.0294699 [74] valid_0's rmse: 0.0294088 [75] valid_0's rmse: 0.0293493 [76] valid_0's rmse: 0.0293017 [77] valid_0's rmse: 0.0292553 [78] valid_0's rmse: 0.0292075 [79] valid_0's rmse: 0.029159 [80] valid_0's rmse: 0.0291097 [81] valid_0's rmse: 0.0290698 [82] valid_0's rmse: 0.0289998 [83] valid_0's rmse: 0.0289311 [84] valid_0's rmse: 0.0288581 [85] valid_0's rmse: 0.0288186 [86] valid_0's rmse: 0.0287897 [87] valid_0's rmse: 0.0287282 [88] valid_0's rmse: 0.0287014 [89] valid_0's rmse: 0.0286655 [90] valid_0's rmse: 0.0286396 [91] valid_0's rmse: 0.0286062 [92] valid_0's rmse: 0.0285594 [93] valid_0's rmse: 0.0285224 [94] valid_0's rmse: 0.0284815 [95] valid_0's rmse: 0.0284341 [96] valid_0's rmse: 0.0283903 [97] valid_0's rmse: 0.0283466 [98] valid_0's rmse: 0.0283241 [99] valid_0's rmse: 0.0282824 [100] valid_0's rmse: 0.028227 [101] valid_0's rmse: 0.0281932 [102] valid_0's rmse: 0.0281535 [103] valid_0's rmse: 0.0280806 [104] valid_0's rmse: 0.0280564 [105] valid_0's rmse: 0.0280251 [106] valid_0's rmse: 0.0280038 [107] valid_0's rmse: 0.0279761 [108] valid_0's rmse: 0.0279315 [109] valid_0's rmse: 0.0278758 [110] valid_0's rmse: 0.0278482 [111] valid_0's rmse: 0.0278085 [112] valid_0's rmse: 0.0277661 [113] valid_0's rmse: 0.0277092 [114] valid_0's rmse: 0.0276904 [115] valid_0's rmse: 0.0276367 [116] valid_0's rmse: 0.0276062 [117] valid_0's rmse: 0.0275664 [118] valid_0's rmse: 0.0275288 [119] valid_0's rmse: 0.0274919 [120] valid_0's rmse: 0.0274641 [121] valid_0's rmse: 0.0274399 [122] valid_0's rmse: 0.0274037 [123] valid_0's rmse: 0.0273689 [124] valid_0's rmse: 0.0273147 [125] valid_0's rmse: 0.0272632 [126] valid_0's rmse: 0.0272447 [127] valid_0's rmse: 0.0272158 [128] valid_0's rmse: 0.0271733 [129] valid_0's rmse: 0.0271393 [130] valid_0's rmse: 0.0270881 [131] valid_0's rmse: 0.0270768 [132] valid_0's rmse: 0.0270641 [133] valid_0's rmse: 0.0270439 [134] valid_0's rmse: 0.0270316 [135] valid_0's rmse: 0.0270212 [136] valid_0's rmse: 0.0270004 [137] valid_0's rmse: 0.0269842 [138] valid_0's rmse: 0.0269647 [139] valid_0's rmse: 0.0269451 [140] valid_0's rmse: 0.0269235 [141] valid_0's rmse: 0.0268934 [142] valid_0's rmse: 0.0268604 [143] valid_0's rmse: 0.0268295 [144] valid_0's rmse: 0.0268063 [145] valid_0's rmse: 0.0267889 [146] valid_0's rmse: 0.0267701 [147] valid_0's rmse: 0.0267576 [148] valid_0's rmse: 0.026751 [149] valid_0's rmse: 0.0267348 [150] valid_0's rmse: 0.0267302 [151] valid_0's rmse: 0.0266969 [152] valid_0's rmse: 0.0266743 [153] valid_0's rmse: 0.0266414 [154] valid_0's rmse: 0.0265964 [155] valid_0's rmse: 0.0265676 [156] valid_0's rmse: 0.0265464 [157] valid_0's rmse: 0.0265254 [158] valid_0's rmse: 0.0264972 [159] valid_0's rmse: 0.0264723 [160] valid_0's rmse: 0.0264595 [161] valid_0's rmse: 0.0264331 [162] valid_0's rmse: 0.0263998 [163] valid_0's rmse: 0.0263718 [164] valid_0's rmse: 0.0263317 [165] valid_0's rmse: 0.0263094 [166] valid_0's rmse: 0.0262948 [167] valid_0's rmse: 0.0262666 [168] valid_0's rmse: 0.0262506 [169] valid_0's rmse: 0.0262277 [170] valid_0's rmse: 0.0262051 [171] valid_0's rmse: 0.0261819 [172] valid_0's rmse: 0.0261827 [173] valid_0's rmse: 0.0261831 [174] valid_0's rmse: 0.0261771 [175] valid_0's rmse: 0.0261687 [176] valid_0's rmse: 0.0261259 [177] valid_0's rmse: 0.026084 [178] valid_0's rmse: 0.0260451 [179] valid_0's rmse: 0.0260044 [180] valid_0's rmse: 0.025966 [181] valid_0's rmse: 0.0259332 [182] valid_0's rmse: 0.0259013 [183] valid_0's rmse: 0.0258755 [184] valid_0's rmse: 0.0258449 [185] valid_0's rmse: 0.0258152 [186] valid_0's rmse: 0.0257944 [187] valid_0's rmse: 0.025773 [188] valid_0's rmse: 0.0257526 [189] valid_0's rmse: 0.0257327 [190] valid_0's rmse: 0.0257139 [191] valid_0's rmse: 0.0257102 [192] valid_0's rmse: 0.025705 [193] valid_0's rmse: 0.0256997 [194] valid_0's rmse: 0.0256948 [195] valid_0's rmse: 0.0256901 [196] valid_0's rmse: 0.0256608 [197] valid_0's rmse: 0.0256519 [198] valid_0's rmse: 0.0256216 [199] valid_0's rmse: 0.0255974 [200] valid_0's rmse: 0.0255765 [201] valid_0's rmse: 0.0255704 [202] valid_0's rmse: 0.0255596 [203] valid_0's rmse: 0.0255437 [204] valid_0's rmse: 0.0255339 [205] valid_0's rmse: 0.0255164 [206] valid_0's rmse: 0.025509 [207] valid_0's rmse: 0.0255068 [208] valid_0's rmse: 0.0255012 [209] valid_0's rmse: 0.0254911 [210] valid_0's rmse: 0.0254842 [211] valid_0's rmse: 0.0254678 [212] valid_0's rmse: 0.0254562 [213] valid_0's rmse: 0.0254432 [214] valid_0's rmse: 0.0254233 [215] valid_0's rmse: 0.0254261 [216] valid_0's rmse: 0.0254124 [217] valid_0's rmse: 0.0253979 [218] valid_0's rmse: 0.0253836 [219] valid_0's rmse: 0.0253722 [220] valid_0's rmse: 0.0253569 [221] valid_0's rmse: 0.0253374 [222] valid_0's rmse: 0.0253137 [223] valid_0's rmse: 0.025305 [224] valid_0's rmse: 0.0252874 [225] valid_0's rmse: 0.0252669 [226] valid_0's rmse: 0.0252575 [227] valid_0's rmse: 0.0252475 [228] valid_0's rmse: 0.0252367 [229] valid_0's rmse: 0.0252271 [230] valid_0's rmse: 0.0252202 [231] valid_0's rmse: 0.0252056 [232] valid_0's rmse: 0.0251975 [233] valid_0's rmse: 0.0251836 [234] valid_0's rmse: 0.0251704 [235] valid_0's rmse: 0.025157 [236] valid_0's rmse: 0.0251498 [237] valid_0's rmse: 0.0251478 [238] valid_0's rmse: 0.0251482 [239] valid_0's rmse: 0.0251432 [240] valid_0's rmse: 0.0251456 [241] valid_0's rmse: 0.025122 [242] valid_0's rmse: 0.0250987 [243] valid_0's rmse: 0.0250986 [244] valid_0's rmse: 0.0250787 [245] valid_0's rmse: 0.0250565 [246] valid_0's rmse: 0.0250366 [247] valid_0's rmse: 0.0250185 [248] valid_0's rmse: 0.0250045 [249] valid_0's rmse: 0.0249908 [250] valid_0's rmse: 0.0249759 [251] valid_0's rmse: 0.0249609 [252] valid_0's rmse: 0.0249443 [253] valid_0's rmse: 0.0249298 [254] valid_0's rmse: 0.024917 [255] valid_0's rmse: 0.0249109 [256] valid_0's rmse: 0.024898 [257] valid_0's rmse: 0.0248832 [258] valid_0's rmse: 0.024868 [259] valid_0's rmse: 0.0248641 [260] valid_0's rmse: 0.0248603 [261] valid_0's rmse: 0.0248412 [262] valid_0's rmse: 0.0248193 [263] valid_0's rmse: 0.0247979 [264] valid_0's rmse: 0.0247795 [265] valid_0's rmse: 0.0247607 [266] valid_0's rmse: 0.0247514 [267] valid_0's rmse: 0.024739 [268] valid_0's rmse: 0.0247273 [269] valid_0's rmse: 0.02472 [270] valid_0's rmse: 0.0247194 [271] valid_0's rmse: 0.0247061 [272] valid_0's rmse: 0.0246918 [273] valid_0's rmse: 0.0246791 [274] valid_0's rmse: 0.024662 [275] valid_0's rmse: 0.024653 [276] valid_0's rmse: 0.0246436 [277] valid_0's rmse: 0.0246298 [278] valid_0's rmse: 0.0246208 [279] valid_0's rmse: 0.0246118 [280] valid_0's rmse: 0.0245976 [281] valid_0's rmse: 0.0245997 [282] valid_0's rmse: 0.0246013 [283] valid_0's rmse: 0.0245843 [284] valid_0's rmse: 0.024582 [285] valid_0's rmse: 0.0245842 [286] valid_0's rmse: 0.0245838 [287] valid_0's rmse: 0.0245806 [288] valid_0's rmse: 0.0245793 [289] valid_0's rmse: 0.0245811 [290] valid_0's rmse: 0.0245856 [291] valid_0's rmse: 0.0245711 [292] valid_0's rmse: 0.0245609 [293] valid_0's rmse: 0.0245539 [294] valid_0's rmse: 0.0245484 [295] valid_0's rmse: 0.0245374 [296] valid_0's rmse: 0.0245338 [297] valid_0's rmse: 0.0245291 [298] valid_0's rmse: 0.0245297 [299] valid_0's rmse: 0.0245264 [300] valid_0's rmse: 0.0245232 [301] valid_0's rmse: 0.0245121 [302] valid_0's rmse: 0.0245017 [303] valid_0's rmse: 0.0244933 [304] valid_0's rmse: 0.0244827 [305] valid_0's rmse: 0.0244746 [306] valid_0's rmse: 0.0244711 [307] valid_0's rmse: 0.0244666 [308] valid_0's rmse: 0.0244632 [309] valid_0's rmse: 0.024459 [310] valid_0's rmse: 0.0244469 [311] valid_0's rmse: 0.024441 [312] valid_0's rmse: 0.0244331 [313] valid_0's rmse: 0.0244258 [314] valid_0's rmse: 0.0244196 [315] valid_0's rmse: 0.0244152 [316] valid_0's rmse: 0.0244001 [317] valid_0's rmse: 0.0243892 [318] valid_0's rmse: 0.0243732 [319] valid_0's rmse: 0.024361 [320] valid_0's rmse: 0.0243533 [321] valid_0's rmse: 0.0243482 [322] valid_0's rmse: 0.0243482 [323] valid_0's rmse: 0.0243467 [324] valid_0's rmse: 0.0243414 [325] valid_0's rmse: 0.0243344 [326] valid_0's rmse: 0.0243312 [327] valid_0's rmse: 0.0243232 [328] valid_0's rmse: 0.0243171 [329] valid_0's rmse: 0.0243081 [330] valid_0's rmse: 0.0243013 [331] valid_0's rmse: 0.0243076 [332] valid_0's rmse: 0.0243055 [333] valid_0's rmse: 0.0243036 [334] valid_0's rmse: 0.0243086 [335] valid_0's rmse: 0.0243114 [336] valid_0's rmse: 0.024314 [337] valid_0's rmse: 0.0243133 [338] valid_0's rmse: 0.0243138 [339] valid_0's rmse: 0.0243095 [340] valid_0's rmse: 0.024309 [341] valid_0's rmse: 0.0242939 [342] valid_0's rmse: 0.0242814 [343] valid_0's rmse: 0.0242711 [344] valid_0's rmse: 0.02426 [345] valid_0's rmse: 0.0242457 [346] valid_0's rmse: 0.0242411 [347] valid_0's rmse: 0.024246 [348] valid_0's rmse: 0.0242413 [349] valid_0's rmse: 0.0242367 [350] valid_0's rmse: 0.024233 [351] valid_0's rmse: 0.0242264 [352] valid_0's rmse: 0.0242207 [353] valid_0's rmse: 0.0242034 [354] valid_0's rmse: 0.0241972 [355] valid_0's rmse: 0.0241909 [356] valid_0's rmse: 0.0241854 [357] valid_0's rmse: 0.0241721 [358] valid_0's rmse: 0.024171 [359] valid_0's rmse: 0.0241659 [360] valid_0's rmse: 0.0241622 [361] valid_0's rmse: 0.0241609 [362] valid_0's rmse: 0.0241598 [363] valid_0's rmse: 0.0241542 [364] valid_0's rmse: 0.024147 [365] valid_0's rmse: 0.0241429 [366] valid_0's rmse: 0.0241384 [367] valid_0's rmse: 0.0241319 [368] valid_0's rmse: 0.0241298 [369] valid_0's rmse: 0.0241277 [370] valid_0's rmse: 0.0241209 [371] valid_0's rmse: 0.0241138 [372] valid_0's rmse: 0.0241083 [373] valid_0's rmse: 0.0241056 [374] valid_0's rmse: 0.0240979 [375] valid_0's rmse: 0.0240917 [376] valid_0's rmse: 0.0240815 [377] valid_0's rmse: 0.0240701 [378] valid_0's rmse: 0.0240618 [379] valid_0's rmse: 0.0240512 [380] valid_0's rmse: 0.0240431 [381] valid_0's rmse: 0.0240451 [382] valid_0's rmse: 0.0240391 [383] valid_0's rmse: 0.0240355 [384] valid_0's rmse: 0.0240338 [385] valid_0's rmse: 0.0240274 [386] valid_0's rmse: 0.0240288 [387] valid_0's rmse: 0.0240293 [388] valid_0's rmse: 0.0240304 [389] valid_0's rmse: 0.0240307 [390] valid_0's rmse: 0.0240326 [391] valid_0's rmse: 0.0240365 [392] valid_0's rmse: 0.0240393 [393] valid_0's rmse: 0.0240363 [394] valid_0's rmse: 0.0240391 [395] valid_0's rmse: 0.0240362 [396] valid_0's rmse: 0.024031 [397] valid_0's rmse: 0.0240299 [398] valid_0's rmse: 0.0240285 [399] valid_0's rmse: 0.0240235 [400] valid_0's rmse: 0.0240225 [401] valid_0's rmse: 0.0240136 [402] valid_0's rmse: 0.0240105 [403] valid_0's rmse: 0.0240043 [404] valid_0's rmse: 0.0239983 [405] valid_0's rmse: 0.0239935 [406] valid_0's rmse: 0.0239919 [407] valid_0's rmse: 0.0239874 [408] valid_0's rmse: 0.0239845 [409] valid_0's rmse: 0.0239855 [410] valid_0's rmse: 0.0239827 [411] valid_0's rmse: 0.023985 [412] valid_0's rmse: 0.0239874 [413] valid_0's rmse: 0.0239847 [414] valid_0's rmse: 0.0239744 [415] valid_0's rmse: 0.0239761 [416] valid_0's rmse: 0.0239762 [417] valid_0's rmse: 0.0239748 [418] valid_0's rmse: 0.0239737 [419] valid_0's rmse: 0.0239707 [420] valid_0's rmse: 0.0239652 [421] valid_0's rmse: 0.023964 [422] valid_0's rmse: 0.0239644 [423] valid_0's rmse: 0.0239657 [424] valid_0's rmse: 0.0239646 [425] valid_0's rmse: 0.0239603 [426] valid_0's rmse: 0.0239571 [427] valid_0's rmse: 0.0239556 [428] valid_0's rmse: 0.023954 [429] valid_0's rmse: 0.0239533 [430] valid_0's rmse: 0.0239492 [431] valid_0's rmse: 0.023952 [432] valid_0's rmse: 0.0239514 [433] valid_0's rmse: 0.0239484 [434] valid_0's rmse: 0.0239519 [435] valid_0's rmse: 0.0239532 [436] valid_0's rmse: 0.023957 [437] valid_0's rmse: 0.0239521 [438] valid_0's rmse: 0.023949 [439] valid_0's rmse: 0.023945 [440] valid_0's rmse: 0.0239412 [441] valid_0's rmse: 0.0239376 [442] valid_0's rmse: 0.0239317 [443] valid_0's rmse: 0.0239337 [444] valid_0's rmse: 0.0239286 [445] valid_0's rmse: 0.0239233 [446] valid_0's rmse: 0.0239192 [447] valid_0's rmse: 0.0239162 [448] valid_0's rmse: 0.0239115 [449] valid_0's rmse: 0.0239118 [450] valid_0's rmse: 0.0239079 [451] valid_0's rmse: 0.0239096 [452] valid_0's rmse: 0.0239117 [453] valid_0's rmse: 0.0239158 [454] valid_0's rmse: 0.023916 [455] valid_0's rmse: 0.0239156 [456] valid_0's rmse: 0.0239162 [457] valid_0's rmse: 0.0239135 [458] valid_0's rmse: 0.0239121 [459] valid_0's rmse: 0.0239084 [460] valid_0's rmse: 0.0239069 [461] valid_0's rmse: 0.0239055 [462] valid_0's rmse: 0.0239052 [463] valid_0's rmse: 0.0239013 [464] valid_0's rmse: 0.0239009 [465] valid_0's rmse: 0.0239033 [466] valid_0's rmse: 0.0239029 [467] valid_0's rmse: 0.0239066 [468] valid_0's rmse: 0.0239107 [469] valid_0's rmse: 0.023904 [470] valid_0's rmse: 0.0239081 [471] valid_0's rmse: 0.0239038 [472] valid_0's rmse: 0.0238995 [473] valid_0's rmse: 0.0238995 [474] valid_0's rmse: 0.0238973 [475] valid_0's rmse: 0.0238976 [476] valid_0's rmse: 0.0238905 [477] valid_0's rmse: 0.0238839 [478] valid_0's rmse: 0.0238772 [479] valid_0's rmse: 0.0238707 [480] valid_0's rmse: 0.0238649 [481] valid_0's rmse: 0.0238679 [482] valid_0's rmse: 0.0238709 [483] valid_0's rmse: 0.023873 [484] valid_0's rmse: 0.0238779 [485] valid_0's rmse: 0.0238804 [486] valid_0's rmse: 0.0238815 [487] valid_0's rmse: 0.023884 [488] valid_0's rmse: 0.023886 [489] valid_0's rmse: 0.0238828 [490] valid_0's rmse: 0.0238858 [491] valid_0's rmse: 0.0238856 [492] valid_0's rmse: 0.0238792 [493] valid_0's rmse: 0.0238776 [494] valid_0's rmse: 0.023879 [495] valid_0's rmse: 0.0238802 [496] valid_0's rmse: 0.0238852 [497] valid_0's rmse: 0.0238801 [498] valid_0's rmse: 0.0238723 [499] valid_0's rmse: 0.0238779 [500] valid_0's rmse: 0.0238718 [501] valid_0's rmse: 0.0238677 [502] valid_0's rmse: 0.0238637 [503] valid_0's rmse: 0.0238597 [504] valid_0's rmse: 0.0238519 [505] valid_0's rmse: 0.0238441 [506] valid_0's rmse: 0.0238405 [507] valid_0's rmse: 0.0238425 [508] valid_0's rmse: 0.0238443 [509] valid_0's rmse: 0.0238466 [510] valid_0's rmse: 0.0238484 [511] valid_0's rmse: 0.0238456 [512] valid_0's rmse: 0.0238436 [513] valid_0's rmse: 0.0238409 [514] valid_0's rmse: 0.0238361 [515] valid_0's rmse: 0.0238312 [516] valid_0's rmse: 0.0238255 [517] valid_0's rmse: 0.0238274 [518] valid_0's rmse: 0.0238291 [519] valid_0's rmse: 0.0238299 [520] valid_0's rmse: 0.023826 [521] valid_0's rmse: 0.023821 [522] valid_0's rmse: 0.0238205 [523] valid_0's rmse: 0.0238157 [524] valid_0's rmse: 0.0238112 [525] valid_0's rmse: 0.0238057 [526] valid_0's rmse: 0.0237963 [527] valid_0's rmse: 0.0237898 [528] valid_0's rmse: 0.0237887 [529] valid_0's rmse: 0.0237883 [530] valid_0's rmse: 0.0237876 [531] valid_0's rmse: 0.0237791 [532] valid_0's rmse: 0.0237714 [533] valid_0's rmse: 0.0237664 [534] valid_0's rmse: 0.0237582 [535] valid_0's rmse: 0.0237576 [536] valid_0's rmse: 0.0237524 [537] valid_0's rmse: 0.0237463 [538] valid_0's rmse: 0.023739 [539] valid_0's rmse: 0.0237339 [540] valid_0's rmse: 0.0237297 [541] valid_0's rmse: 0.0237259 [542] valid_0's rmse: 0.0237235 [543] valid_0's rmse: 0.0237196 [544] valid_0's rmse: 0.023717 [545] valid_0's rmse: 0.0237148 [546] valid_0's rmse: 0.0237126 [547] valid_0's rmse: 0.0237099 [548] valid_0's rmse: 0.0237077 [549] valid_0's rmse: 0.0237091 [550] valid_0's rmse: 0.0237079 [551] valid_0's rmse: 0.0237111 [552] valid_0's rmse: 0.0237143 [553] valid_0's rmse: 0.0237081 [554] valid_0's rmse: 0.0237075 [555] valid_0's rmse: 0.0237112 [556] valid_0's rmse: 0.0237092 [557] valid_0's rmse: 0.0237085 [558] valid_0's rmse: 0.0237046 [559] valid_0's rmse: 0.0237071 [560] valid_0's rmse: 0.0237038 [561] valid_0's rmse: 0.0237053 [562] valid_0's rmse: 0.0237058 [563] valid_0's rmse: 0.0237081 [564] valid_0's rmse: 0.0237097 [565] valid_0's rmse: 0.0237125 [566] valid_0's rmse: 0.02371 [567] valid_0's rmse: 0.0237057 [568] valid_0's rmse: 0.0237037 [569] valid_0's rmse: 0.0236994 [570] valid_0's rmse: 0.0236973 [571] valid_0's rmse: 0.0237007 [572] valid_0's rmse: 0.0237056 [573] valid_0's rmse: 0.0237081 [574] valid_0's rmse: 0.0237099 [575] valid_0's rmse: 0.0237144 [576] valid_0's rmse: 0.0237134 [577] valid_0's rmse: 0.0237125 [578] valid_0's rmse: 0.023717 [579] valid_0's rmse: 0.0237166 [580] valid_0's rmse: 0.023721 [581] valid_0's rmse: 0.0237188 [582] valid_0's rmse: 0.0237185 [583] valid_0's rmse: 0.0237182 [584] valid_0's rmse: 0.0237154 [585] valid_0's rmse: 0.0237144 [586] valid_0's rmse: 0.0237223 [587] valid_0's rmse: 0.0237258 [588] valid_0's rmse: 0.0237284 [589] valid_0's rmse: 0.0237351 [590] valid_0's rmse: 0.0237396 [591] valid_0's rmse: 0.0237399 [592] valid_0's rmse: 0.0237394 [593] valid_0's rmse: 0.0237347 [594] valid_0's rmse: 0.0237336 [595] valid_0's rmse: 0.0237299 [596] valid_0's rmse: 0.0237252 [597] valid_0's rmse: 0.0237205 [598] valid_0's rmse: 0.0237181 [599] valid_0's rmse: 0.0237155 [600] valid_0's rmse: 0.0237113 [601] valid_0's rmse: 0.0237135 [602] valid_0's rmse: 0.0237156 [603] valid_0's rmse: 0.0237178 [604] valid_0's rmse: 0.0237191 [605] valid_0's rmse: 0.0237146 [606] valid_0's rmse: 0.0237114 [607] valid_0's rmse: 0.0237142 [608] valid_0's rmse: 0.023715 [609] valid_0's rmse: 0.0237133 [610] valid_0's rmse: 0.0237139 [611] valid_0's rmse: 0.0237155 [612] valid_0's rmse: 0.0237169 [613] valid_0's rmse: 0.0237189 [614] valid_0's rmse: 0.0237202 [615] valid_0's rmse: 0.0237216 [616] valid_0's rmse: 0.0237216 [617] valid_0's rmse: 0.0237165 [618] valid_0's rmse: 0.0237185 [619] valid_0's rmse: 0.0237152 [620] valid_0's rmse: 0.0237125 [621] valid_0's rmse: 0.0237107 [622] valid_0's rmse: 0.0237079 [623] valid_0's rmse: 0.0237081 [624] valid_0's rmse: 0.0237024 [625] valid_0's rmse: 0.0237027 [626] valid_0's rmse: 0.0237078 [627] valid_0's rmse: 0.0237062 [628] valid_0's rmse: 0.0237076 [629] valid_0's rmse: 0.0237107 [630] valid_0's rmse: 0.0237141 [631] valid_0's rmse: 0.02372 [632] valid_0's rmse: 0.0237201 [633] valid_0's rmse: 0.0237182 [634] valid_0's rmse: 0.023717 [635] valid_0's rmse: 0.023711 [636] valid_0's rmse: 0.0237045 [637] valid_0's rmse: 0.0237 [638] valid_0's rmse: 0.0236946 [639] valid_0's rmse: 0.0236963 [640] valid_0's rmse: 0.023692 [641] valid_0's rmse: 0.0236891 [642] valid_0's rmse: 0.0236864 [643] valid_0's rmse: 0.0236813 [644] valid_0's rmse: 0.0236781 [645] valid_0's rmse: 0.0236752 [646] valid_0's rmse: 0.0236743 [647] valid_0's rmse: 0.0236729 [648] valid_0's rmse: 0.0236707 [649] valid_0's rmse: 0.0236687 [650] valid_0's rmse: 0.0236681 [651] valid_0's rmse: 0.0236664 [652] valid_0's rmse: 0.0236637 [653] valid_0's rmse: 0.0236597 [654] valid_0's rmse: 0.0236582 [655] valid_0's rmse: 0.0236561 [656] valid_0's rmse: 0.0236593 [657] valid_0's rmse: 0.0236551 [658] valid_0's rmse: 0.0236587 [659] valid_0's rmse: 0.0236519 [660] valid_0's rmse: 0.0236453 [661] valid_0's rmse: 0.0236416 [662] valid_0's rmse: 0.023636 [663] valid_0's rmse: 0.0236345 [664] valid_0's rmse: 0.0236303 [665] valid_0's rmse: 0.0236267 [666] valid_0's rmse: 0.0236234 [667] valid_0's rmse: 0.0236212 [668] valid_0's rmse: 0.0236192 [669] valid_0's rmse: 0.023616 [670] valid_0's rmse: 0.0236154 [671] valid_0's rmse: 0.0236192 [672] valid_0's rmse: 0.0236205 [673] valid_0's rmse: 0.0236209 [674] valid_0's rmse: 0.0236217 [675] valid_0's rmse: 0.0236231 [676] valid_0's rmse: 0.0236199 [677] valid_0's rmse: 0.0236168 [678] valid_0's rmse: 0.0236158 [679] valid_0's rmse: 0.0236132 [680] valid_0's rmse: 0.023615 [681] valid_0's rmse: 0.0236104 [682] valid_0's rmse: 0.0236094 [683] valid_0's rmse: 0.023608 [684] valid_0's rmse: 0.0236061 [685] valid_0's rmse: 0.0236054 [686] valid_0's rmse: 0.0236006 [687] valid_0's rmse: 0.0235964 [688] valid_0's rmse: 0.0235926 [689] valid_0's rmse: 0.0235928 [690] valid_0's rmse: 0.0235919 [691] valid_0's rmse: 0.0235965 [692] valid_0's rmse: 0.023596 [693] valid_0's rmse: 0.0235979 [694] valid_0's rmse: 0.0235969 [695] valid_0's rmse: 0.0235965 [696] valid_0's rmse: 0.0235954 [697] valid_0's rmse: 0.0235889 [698] valid_0's rmse: 0.0235825 [699] valid_0's rmse: 0.0235817 [700] valid_0's rmse: 0.0235755 [701] valid_0's rmse: 0.0235773 [702] valid_0's rmse: 0.0235708 [703] valid_0's rmse: 0.0235742 [704] valid_0's rmse: 0.0235722 [705] valid_0's rmse: 0.0235752 [706] valid_0's rmse: 0.0235732 [707] valid_0's rmse: 0.0235712 [708] valid_0's rmse: 0.023572 [709] valid_0's rmse: 0.0235776 [710] valid_0's rmse: 0.0235757 [711] valid_0's rmse: 0.0235846 [712] valid_0's rmse: 0.0235865 [713] valid_0's rmse: 0.0235889 [714] valid_0's rmse: 0.0235902 [715] valid_0's rmse: 0.0235915 [716] valid_0's rmse: 0.0235899 [717] valid_0's rmse: 0.0235843 [718] valid_0's rmse: 0.0235871 [719] valid_0's rmse: 0.0235815 [720] valid_0's rmse: 0.0235834 [721] valid_0's rmse: 0.0235796 [722] valid_0's rmse: 0.0235709 [723] valid_0's rmse: 0.0235679 [724] valid_0's rmse: 0.0235597 [725] valid_0's rmse: 0.023558 [726] valid_0's rmse: 0.0235521 [727] valid_0's rmse: 0.0235506 [728] valid_0's rmse: 0.0235456 [729] valid_0's rmse: 0.0235434 [730] valid_0's rmse: 0.023542 [731] valid_0's rmse: 0.0235465 [732] valid_0's rmse: 0.0235518 [733] valid_0's rmse: 0.0235577 [734] valid_0's rmse: 0.0235628 [735] valid_0's rmse: 0.023568 [736] valid_0's rmse: 0.023567 [737] valid_0's rmse: 0.0235687 [738] valid_0's rmse: 0.023567 [739] valid_0's rmse: 0.0235689 [740] valid_0's rmse: 0.0235683 [741] valid_0's rmse: 0.0235687 [742] valid_0's rmse: 0.0235667 [743] valid_0's rmse: 0.023565 [744] valid_0's rmse: 0.0235647 [745] valid_0's rmse: 0.0235639 [746] valid_0's rmse: 0.0235626 [747] valid_0's rmse: 0.0235621 [748] valid_0's rmse: 0.0235655 [749] valid_0's rmse: 0.0235651 [750] valid_0's rmse: 0.0235626 [751] valid_0's rmse: 0.0235604 [752] valid_0's rmse: 0.0235554 [753] valid_0's rmse: 0.023553 [754] valid_0's rmse: 0.0235507 [755] valid_0's rmse: 0.0235485 [756] valid_0's rmse: 0.0235475 [757] valid_0's rmse: 0.0235471 [758] valid_0's rmse: 0.023547 [759] valid_0's rmse: 0.023546 [760] valid_0's rmse: 0.023548 [761] valid_0's rmse: 0.0235404 [762] valid_0's rmse: 0.0235322 [763] valid_0's rmse: 0.0235255 [764] valid_0's rmse: 0.0235206 [765] valid_0's rmse: 0.0235189 [766] valid_0's rmse: 0.0235161 [767] valid_0's rmse: 0.0235131 [768] valid_0's rmse: 0.0235113 [769] valid_0's rmse: 0.0235116 [770] valid_0's rmse: 0.0235096 [771] valid_0's rmse: 0.0235095 [772] valid_0's rmse: 0.0235065 [773] valid_0's rmse: 0.0235049 [774] valid_0's rmse: 0.023504 [775] valid_0's rmse: 0.0235031 [776] valid_0's rmse: 0.0235041 [777] valid_0's rmse: 0.0235034 [778] valid_0's rmse: 0.0235044 [779] valid_0's rmse: 0.0235023 [780] valid_0's rmse: 0.0235033 [781] valid_0's rmse: 0.023505 [782] valid_0's rmse: 0.0235067 [783] valid_0's rmse: 0.0235069 [784] valid_0's rmse: 0.023507 [785] valid_0's rmse: 0.0235084 [786] valid_0's rmse: 0.0235073 [787] valid_0's rmse: 0.0235103 [788] valid_0's rmse: 0.0235107 [789] valid_0's rmse: 0.0235102 [790] valid_0's rmse: 0.0235127 [791] valid_0's rmse: 0.0235112 [792] valid_0's rmse: 0.0235102 [793] valid_0's rmse: 0.0235093 [794] valid_0's rmse: 0.0235079 [795] valid_0's rmse: 0.0235067 [796] valid_0's rmse: 0.0235103 [797] valid_0's rmse: 0.0235153 [798] valid_0's rmse: 0.0235183 [799] valid_0's rmse: 0.0235219 [800] valid_0's rmse: 0.0235219 [801] valid_0's rmse: 0.0235216 [802] valid_0's rmse: 0.0235214 [803] valid_0's rmse: 0.0235148 [804] valid_0's rmse: 0.0235062 [805] valid_0's rmse: 0.0235034 [806] valid_0's rmse: 0.0235047 [807] valid_0's rmse: 0.0235054 [808] valid_0's rmse: 0.0235062 [809] valid_0's rmse: 0.023506 [810] valid_0's rmse: 0.0235021 [811] valid_0's rmse: 0.0235045 [812] valid_0's rmse: 0.0235052 [813] valid_0's rmse: 0.023506 [814] valid_0's rmse: 0.0235077 [815] valid_0's rmse: 0.0235097 [816] valid_0's rmse: 0.0235131 [817] valid_0's rmse: 0.0235155 [818] valid_0's rmse: 0.0235176 [819] valid_0's rmse: 0.0235196 [820] valid_0's rmse: 0.0235238 [821] valid_0's rmse: 0.0235239 [822] valid_0's rmse: 0.0235244 [823] valid_0's rmse: 0.0235239 [824] valid_0's rmse: 0.0235241 [825] valid_0's rmse: 0.0235212 [826] valid_0's rmse: 0.0235226 [827] valid_0's rmse: 0.023523 [828] valid_0's rmse: 0.023524 [829] valid_0's rmse: 0.0235217 [830] valid_0's rmse: 0.0235221 [831] valid_0's rmse: 0.023523 [832] valid_0's rmse: 0.023519 [833] valid_0's rmse: 0.0235185 [834] valid_0's rmse: 0.0235189 [835] valid_0's rmse: 0.0235118 [836] valid_0's rmse: 0.0235092 [837] valid_0's rmse: 0.0235095 [838] valid_0's rmse: 0.0235097 [839] valid_0's rmse: 0.02351 [840] valid_0's rmse: 0.0235108 [841] valid_0's rmse: 0.0235141 [842] valid_0's rmse: 0.0235179 [843] valid_0's rmse: 0.0235229 [844] valid_0's rmse: 0.0235268 [845] valid_0's rmse: 0.0235272 [846] valid_0's rmse: 0.0235265 [847] valid_0's rmse: 0.0235274 [848] valid_0's rmse: 0.0235266 [849] valid_0's rmse: 0.0235251 [850] valid_0's rmse: 0.0235253 [851] valid_0's rmse: 0.0235213 [852] valid_0's rmse: 0.023517 [853] valid_0's rmse: 0.0235131 [854] valid_0's rmse: 0.0235091 [855] valid_0's rmse: 0.0235049 [856] valid_0's rmse: 0.0235052 [857] valid_0's rmse: 0.0235 [858] valid_0's rmse: 0.0235013 [859] valid_0's rmse: 0.0235023 [860] valid_0's rmse: 0.0235047 [861] valid_0's rmse: 0.023503 [862] valid_0's rmse: 0.0234994 [863] valid_0's rmse: 0.0234981 [864] valid_0's rmse: 0.0234936 [865] valid_0's rmse: 0.0234924 [866] valid_0's rmse: 0.0234934 [867] valid_0's rmse: 0.0234969 [868] valid_0's rmse: 0.0234978 [869] valid_0's rmse: 0.0235012 [870] valid_0's rmse: 0.0235047 [871] valid_0's rmse: 0.0235039 [872] valid_0's rmse: 0.0235025 [873] valid_0's rmse: 0.0235049 [874] valid_0's rmse: 0.0235041 [875] valid_0's rmse: 0.0235059 [876] valid_0's rmse: 0.0235107 [877] valid_0's rmse: 0.0235122 [878] valid_0's rmse: 0.0235171 [879] valid_0's rmse: 0.0235214 [880] valid_0's rmse: 0.0235213 [881] valid_0's rmse: 0.023521 [882] valid_0's rmse: 0.0235209 [883] valid_0's rmse: 0.0235202 [884] valid_0's rmse: 0.0235197 [885] valid_0's rmse: 0.0235206 [886] valid_0's rmse: 0.0235238 [887] valid_0's rmse: 0.0235254 [888] valid_0's rmse: 0.0235276 [889] valid_0's rmse: 0.0235311 [890] valid_0's rmse: 0.0235339 [891] valid_0's rmse: 0.0235356 [892] valid_0's rmse: 0.0235324 [893] valid_0's rmse: 0.0235339 [894] valid_0's rmse: 0.0235337 [895] valid_0's rmse: 0.0235358 [896] valid_0's rmse: 0.0235353 [897] valid_0's rmse: 0.0235357 [898] valid_0's rmse: 0.0235341 [899] valid_0's rmse: 0.0235337 [900] valid_0's rmse: 0.023535 [901] valid_0's rmse: 0.0235356 [902] valid_0's rmse: 0.0235366 [903] valid_0's rmse: 0.0235374 [904] valid_0's rmse: 0.023538 [905] valid_0's rmse: 0.0235402 [906] valid_0's rmse: 0.023542 [907] valid_0's rmse: 0.0235439 [908] valid_0's rmse: 0.0235459 [909] valid_0's rmse: 0.0235494 [910] valid_0's rmse: 0.0235523 [911] valid_0's rmse: 0.0235503 [912] valid_0's rmse: 0.0235494 [913] valid_0's rmse: 0.0235473 [914] valid_0's rmse: 0.0235462 [915] valid_0's rmse: 0.0235449 [916] valid_0's rmse: 0.0235459 [917] valid_0's rmse: 0.023544 [918] valid_0's rmse: 0.0235448 [919] valid_0's rmse: 0.023543 [920] valid_0's rmse: 0.0235422 [921] valid_0's rmse: 0.0235431 [922] valid_0's rmse: 0.023542 [923] valid_0's rmse: 0.0235415 [924] valid_0's rmse: 0.0235385 [925] valid_0's rmse: 0.0235385 [926] valid_0's rmse: 0.0235357 [927] valid_0's rmse: 0.0235344 [928] valid_0's rmse: 0.0235349 [929] valid_0's rmse: 0.0235341 [930] valid_0's rmse: 0.0235303 [931] valid_0's rmse: 0.0235341 [932] valid_0's rmse: 0.023536 [933] valid_0's rmse: 0.0235349 [934] valid_0's rmse: 0.0235368 [935] valid_0's rmse: 0.0235389 [936] valid_0's rmse: 0.0235382 [937] valid_0's rmse: 0.0235344 [938] valid_0's rmse: 0.0235317 [939] valid_0's rmse: 0.0235276 [940] valid_0's rmse: 0.0235269 [941] valid_0's rmse: 0.0235273 [942] valid_0's rmse: 0.0235316 [943] valid_0's rmse: 0.0235336 [944] valid_0's rmse: 0.0235346 [945] valid_0's rmse: 0.0235367 [946] valid_0's rmse: 0.023537 [947] valid_0's rmse: 0.023538 [948] valid_0's rmse: 0.023539 [949] valid_0's rmse: 0.0235394 [950] valid_0's rmse: 0.0235403 [951] valid_0's rmse: 0.0235336 [952] valid_0's rmse: 0.0235269 [953] valid_0's rmse: 0.0235204 [954] valid_0's rmse: 0.0235217 [955] valid_0's rmse: 0.0235153 [956] valid_0's rmse: 0.0235105 [957] valid_0's rmse: 0.023508 [958] valid_0's rmse: 0.0235124 [959] valid_0's rmse: 0.023517 [960] valid_0's rmse: 0.0235178 [961] valid_0's rmse: 0.0235156 [962] valid_0's rmse: 0.0235136 [963] valid_0's rmse: 0.0235115 [964] valid_0's rmse: 0.023511 [965] valid_0's rmse: 0.023509 Early stopping, best iteration is: [865] valid_0's rmse: 0.0234924
In [15]:
y_pred = np.expm1(np.expm1(gbm.predict(X_test))) y_true = np.expm1(np.expm1(Y_test))
In [16]:
MSE = mean_squared_error(y_true, y_pred) RMSE = np.sqrt(mean_squared_error(y_true, y_pred)) MAE = mean_absolute_error(y_true, y_pred) MAPE = mean_absolute_percentage_error(y_true, y_pred) R_2 = r2_score(y_true, y_pred) print('MSE:', format(MSE, '.2E')) print('RMSE:', RMSE) print('MAE:', MAE) print('MAPE:', MAPE) print('R_2:', R_2) #R方为负就说明拟合效果比平均值差
MSE: 5.74E+03 RMSE: 75.76181372320752 MAE: 54.761485656796516 MAPE: 0.11813067374720394 R_2: 0.6942490637163895
In [13]:
MSE = mean_squared_error(y_true, y_pred) RMSE = np.sqrt(mean_squared_error(y_true, y_pred)) MAE = mean_absolute_error(y_true, y_pred) MAPE = mean_absolute_percentage_error(y_true, y_pred) R_2 = r2_score(y_true, y_pred) print('MSE:', format(MSE, '.2E')) print('RMSE:', RMSE) print('MAE:', MAE) print('MAPE:', MAPE) print('R_2:', R_2) #R方为负就说明拟合效果比平均值差
MSE: 5.55E+02 RMSE: 23.55984288394111 MAE: 17.576455710512402 MAPE: 0.03501471248540124 R_2: 0.9704326876428987
In [20]:
import seaborn as sns
In [23]:
feature_importance = pd.DataFrame() feature_importance['fea_name'] = feature_cols feature_importance['fea_imp'] = gbm.feature_importance() feature_importance = feature_importance.sort_values('fea_imp', ascending=False) plt.figure(figsize=[40, 20]) ax = sns.barplot(x=feature_importance['fea_name'], y=feature_importance['fea_imp']) ax.set_xticklabels(labels=feature_cols, rotation=45, fontsize=6) ax.set_yticklabels(labels=[0, 2000, 4000, 6000, 8000, 10000, 12000, 14000, 16000], fontsize=6) plt.xlabel('fea_name', fontsize=18) plt.ylabel('fea_imp', fontsize=18) plt.tight_layout() plt.savefig('./figure/特征重要性.png')
C:\Users\zhaojh\AppData\Local\Temp\ipykernel_1488\3332664617.py:9: UserWarning: FixedFormatter should only be used together with FixedLocator ax.set_yticklabels(labels=[0, 2000, 4000, 6000, 8000, 10000, 12000, 14000, 16000], fontsize=6)
In [18]:
import seaborn as sns
In [22]:
from sklearn.preprocessing import MinMaxScaler
In [ ]: