TypeError: ufunc 'isnan' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

[kelvin-ksl]

05-04 20:09:07 E hypernets.e._experiment.py 106 - ExperimentID:[HyperTS_de1045f0eeffff25abcbae84d1758aed] - None: ufunc 'isnan' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''.

How resolve?

[zhangxjohn]

I can not understand your descirption. Can you give me more detials?

[kelvin-ksl]

I'm using 2 datasets, one for training and the other for testing, I follow example 14 (anomaly_detection_experiment) but in the temporal_train_test_split section I put the following syntax:

train_data = train_user2 test_data = df

other than that the code is identical to the example

*Error occurs in make_experiment

[zhangxjohn]

Ok, please first check the data format of train_dataand test_data , 'anomaly'should not in their columns. In addition, what is the train_user2?

[zhangxjohn]

If you do not fix the code in the section temporal_train_test_split, anyError occurs?

[kelvin-ksl]

I'm assigning the anomaly column to the ground_truth variable as shown in the example.

train_user2 is my training data, in it I didn't remove the anomaly column.

[kelvin-ksl]

no error occurs before , only in the make experience section.

If you wish, I can share the notebook with the data I am using, I will only ask for your email to send it.

[zhangxjohn]

I am open, you can.

[kelvin-ksl]

sent.

[zhangxjohn]

Attention, my email is: zhangxjohn@yeah.net