coal_materials/20240123_煤沥青-checkpoint.ipynb at master

606 KiB

Raw Permalink Blame History

In [1]:

import pandas as pd

In [2]:

data = pd.read_excel('./data/20240123/煤沥青数据.xlsx')
data.head()

Out[2]:

	碳源	共碳化物质	加热次数	是否有碳化过程	模板剂种类	模板剂比例	KOH与煤沥青比例	活化温度	升温速率	活化时间	混合方式	比表面积	总孔体积	微孔体积	平均孔径
0	煤沥青	无	1	否	自制氧化钙	1.0	1.0	500	5	2.0	溶剂	908.07	0.40	0.34	1.75
1	煤沥青	无	1	否	自制氧化钙	1.0	0.5	600	5	2.0	溶剂	953.95	0.66	0.35	2.76
2	煤沥青	无	1	否	自制氧化钙	1.0	1.0	600	5	2.0	溶剂	1388.62	0.61	0.53	1.77
3	煤沥青	无	1	否	自制氧化钙	1.0	2.0	600	5	2.0	溶剂	1444.63	0.59	0.55	1.62
4	煤沥青	无	2	是	自制碱式碳酸镁	1.0	1.0	600	5	2.0	溶剂	1020.99	0.45	0.35	1.77

In [3]:

data.shape

Out[3]:

(149, 16)

In [4]:

data.columns

Out[4]:

Index(['碳源', '共碳化物质', '共碳化物/煤沥青', '加热次数', '是否有碳化过程', '模板剂种类', '模板剂比例',
       'KOH与煤沥青比例', '活化温度', '升温速率', '活化时间', '混合方式', '比表面积', '总孔体积', '微孔体积',
       '平均孔径'],
      dtype='object')

In [5]:

data.drop(columns=['碳源'], inplace=True)

In [6]:

object_cols = ['共碳化物质', '是否有碳化过程', '模板剂种类', '混合方式']

In [7]:

data = pd.get_dummies(data, columns=object_cols)

In [8]:

out_cols = ['比表面积', '总孔体积', '微孔体积', '平均孔径']
feature_cols = [x for x in data.columns if x not in out_cols]

In [9]:

train_data = data.reset_index(drop=True)

In [10]:

train_data.shape

Out[10]:

(149, 40)

In [11]:

import xgboost as xgb
from sklearn.metrics import mean_absolute_error, mean_squared_error, mean_absolute_percentage_error, r2_score

In [12]:

from sklearn.model_selection import KFold, train_test_split
kf = KFold(n_splits=5, shuffle=True, random_state=666)

In [13]:

import numpy as np

In [14]:

params_xgb = {"objective": 'reg:squarederror',
             "subsample": 0.9,
             "max_depth": 20,
             "eta": 0.01,
             "colsample_bytree": 0.9,}
num_boost_round = 1000

In [15]:

import matplotlib.pyplot as plt

In [16]:

plt.rcParams["font.sans-serif"] = ["SimHei"]  # 设置字体
plt.rcParams["axes.unicode_minus"] = False  # 正常显示负号

In [18]:

eva_total = list()
index_list = list()
eva_cols = ['MSE', 'RMSE', 'MAE', 'MAPE', 'R2']
for col in out_cols:
    eva_list = list()
    train_data = train_data[~train_data[col].isna()].reset_index(drop=True)
    cur_test = list()
    cur_real = list()
    for (train_index, test_index) in kf.split(train_data):
        train = train_data.loc[train_index]
        valid = train_data.loc[test_index]
        X_train, Y_train = train[feature_cols], train[col]
        X_valid, Y_valid = valid[feature_cols], valid[col]
        dtrain = xgb.DMatrix(X_train, Y_train)
        dvalid = xgb.DMatrix(X_valid, Y_valid)
        watchlist = [(dvalid, 'eval')]
        gb_model = xgb.train(params_xgb, dtrain, num_boost_round, evals=watchlist,
                        early_stopping_rounds=100, verbose_eval=False)
        y_pred = gb_model.predict(xgb.DMatrix(X_valid))
        y_true = Y_valid.values
        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)
        cur_test.extend(y_pred[:7])
        cur_real.extend(y_true[:7])
        print('MSE:', round(MSE, 4), end=', ')
        print('RMSE:', round(RMSE, 4), end=', ')
        print('MAE:', round(MAE, 4), end=', ')
        print('MAPE:', round(MAPE*100, 2), '%', end=', ')
        print('R_2:', round(R_2, 4))  #R方为负就说明拟合效果比平均值差
        eva_list.append([MSE, RMSE, MAE, MAPE, R_2])
    plt.figure(figsize=(12, 8))
    plt.plot(range(len(cur_test)), cur_real, 'o-', label='real')
    plt.plot(range(len(cur_test)), cur_test, '*-', label='pred')
    plt.legend(loc='best')
    plt.title(f'{col}')
    plt.show()
    eva_total.append(np.mean(eva_list, axis=0))
    index_list.append(f"{col}")

MSE: 146052.1781, RMSE: 382.1677, MAE: 335.8072, MAPE: 27.62 %, R_2: 0.3237
MSE: 457536.2053, RMSE: 676.4142, MAE: 524.2504, MAPE: 436.7 %, R_2: 0.4597
MSE: 426986.1964, RMSE: 653.4418, MAE: 517.7005, MAPE: 28.25 %, R_2: 0.1735
MSE: 276509.2691, RMSE: 525.8415, MAE: 387.3172, MAPE: 32.43 %, R_2: 0.4786
MSE: 300204.7099, RMSE: 547.9094, MAE: 395.1222, MAPE: 314.87 %, R_2: 0.3381
MSE: 243884.6623, RMSE: 493.8468, MAE: 382.9586, MAPE: 1077.01 %, R_2: 0.6543
MSE: 380516.2705, RMSE: 616.86, MAE: 528.3397, MAPE: 42.43 %, R_2: 0.294
MSE: 457352.6686, RMSE: 676.2785, MAE: 515.0433, MAPE: 547.78 %, R_2: 0.5355
MSE: 275148.3579, RMSE: 524.5459, MAE: 464.9701, MAPE: 48.3 %, R_2: 0.3033
MSE: 215299.6743, RMSE: 464.004, MAE: 385.4702, MAPE: 20.69 %, R_2: 0.4055

No description has been provided for this image

MSE: 0.0565, RMSE: 0.2377, MAE: 0.1623, MAPE: 30.73 %, R_2: 0.7346
MSE: 0.1443, RMSE: 0.3798, MAE: 0.2874, MAPE: 154.87 %, R_2: 0.801
MSE: 0.3168, RMSE: 0.5628, MAE: 0.4358, MAPE: 43.0 %, R_2: 0.4067
MSE: 0.1148, RMSE: 0.3389, MAE: 0.2794, MAPE: 29.47 %, R_2: 0.597
MSE: 0.1082, RMSE: 0.329, MAE: 0.2451, MAPE: 125.28 %, R_2: 0.3208
MSE: 0.0987, RMSE: 0.3141, MAE: 0.2595, MAPE: 338.62 %, R_2: 0.6563
MSE: 0.1457, RMSE: 0.3817, MAE: 0.2933, MAPE: 40.45 %, R_2: 0.3009
MSE: 0.1538, RMSE: 0.3922, MAE: 0.3011, MAPE: 441.83 %, R_2: 0.4244
MSE: 0.1302, RMSE: 0.3609, MAE: 0.2923, MAPE: 46.21 %, R_2: 0.1553
MSE: 0.0737, RMSE: 0.2715, MAE: 0.2209, MAPE: 22.16 %, R_2: 0.6708

MSE: 0.0334, RMSE: 0.1829, MAE: 0.1388, MAPE: 28.81 %, R_2: -0.1911
MSE: 0.0604, RMSE: 0.2457, MAE: 0.1958, MAPE: 1278.96 %, R_2: 0.6126
MSE: 0.0646, RMSE: 0.2542, MAE: 0.1992, MAPE: 41.7 %, R_2: 0.3841
MSE: 0.0459, RMSE: 0.2142, MAE: 0.153, MAPE: 38.47 %, R_2: 0.563
MSE: 0.0213, RMSE: 0.1459, MAE: 0.1258, MAPE: 216.58 %, R_2: 0.6774
MSE: 0.0332, RMSE: 0.1822, MAE: 0.1545, MAPE: 1355.85 %, R_2: 0.7458
MSE: 0.0534, RMSE: 0.231, MAE: 0.1976, MAPE: 63.72 %, R_2: 0.19
MSE: 0.0217, RMSE: 0.1474, MAE: 0.1131, MAPE: 1044.2 %, R_2: 0.7267
MSE: 0.09, RMSE: 0.3, MAE: 0.261, MAPE: 76.62 %, R_2: 0.058
MSE: 0.0616, RMSE: 0.2482, MAE: 0.1984, MAPE: 44.34 %, R_2: 0.3462

MSE: 0.7362, RMSE: 0.858, MAE: 0.5955, MAPE: 23.28 %, R_2: 0.6539
MSE: 0.6157, RMSE: 0.7847, MAE: 0.4994, MAPE: 15.91 %, R_2: 0.3943
MSE: 0.2767, RMSE: 0.526, MAE: 0.4052, MAPE: 18.6 %, R_2: 0.6936
MSE: 0.5095, RMSE: 0.7138, MAE: 0.4774, MAPE: 16.58 %, R_2: 0.7721
MSE: 2.0145, RMSE: 1.4193, MAE: 0.9327, MAPE: 24.89 %, R_2: 0.0073
MSE: 1.4449, RMSE: 1.202, MAE: 0.5396, MAPE: 13.41 %, R_2: 0.2873
MSE: 0.4101, RMSE: 0.6404, MAE: 0.4024, MAPE: 16.32 %, R_2: -1.8128
MSE: 1.4384, RMSE: 1.1993, MAE: 0.8923, MAPE: 39.29 %, R_2: -0.313
MSE: 0.3805, RMSE: 0.6168, MAE: 0.4343, MAPE: 16.0 %, R_2: 0.6971
MSE: 0.0937, RMSE: 0.3061, MAE: 0.2214, MAPE: 9.25 %, R_2: 0.9435

In [19]:

pd.DataFrame.from_records(eva_total, index=index_list, columns=eva_cols)

Out[19]:

	MSE	RMSE	MAE	MAPE	R2
比表面积	317949.019239	556.130983	443.697934	2.576080	0.396622
总孔体积	0.134278	0.356866	0.277702	1.272616	0.506787
微孔体积	0.048545	0.215159	0.173718	4.189249	0.411268
平均孔径	0.792019	0.826652	0.540024	0.193539	0.232322

In [19]:

pd.DataFrame.from_records(eva_total, index=index_list, columns=eva_cols)

Out[19]:

	MSE	RMSE	MAE	MAPE	R2
比表面积	315103.064636	547.014045	398.015647	0.976955	0.466536
总孔体积	0.127985	0.353757	0.268668	1.179238	0.470801
微孔体积	0.044655	0.207595	0.164963	4.548583	0.536350
平均孔径	0.801410	0.828470	0.533512	0.184272	0.302607

In [ ]:

606 KiB Raw Permalink Blame History

606 KiB

Raw Permalink Blame History