使用Autogluon时出现'TabularDataset'对象无'unique'属性报错

我的代码如下：

target=['A1','A2','A3','A4','A5','H1','H2','H3','H4','H5','Lf']
features=['miuH','rhoH','YH_OC','YH_O2','YH_N']

train_df=TabularDataset('train_df.csv')
test_df=TabularDataset('test_df.csv')

predictor = TabularPredictor(label=target).fit(
    train_df[features+target],
    tuning_data=test_df[features+target],
    time_limit=240
)

数据集片段：

Unnamed: 0    miuH      rhoH  ...             H4            H5        Lf
0           0  2.7500   80000.0  ...   76131.310000  7.699976e+04  0.000680
1           1  1.6250  140000.0  ...   69323.414000  6.419696e+04  0.000704
2           2  3.8750   40000.0  ...   39790.990000  3.989756e+04  0.000382
3           3  1.0625  100000.0  ...       0.000054  2.731176e-12  0.000403
4           4  3.3125  160000.0  ...  111573.800000  1.188784e+05  0.000579

[5 rows x 17 columns]

求各位帮忙排查问题！

内容的提问来源于stack exchange，提问作者Chao Luo