search

hatalis / Quantile_Detector

Qauntile autoregressive neural network for time series anamoly detection.
2 stars 2 forks source link

TypeError: drop() got an unexpected keyword argument 'columns' #1

Open ghost opened 5 years ago

ghost commented 5 years ago

Hello, I tried your code and got this error:

python main.py Using TensorFlow backend. Traceback (most recent call last): File "main.py", line 27, in experiment = split_data(experiment, scaling=MinMaxScaler(), test_size=2/30) # StandardScaler() File "C:\Work\code\processing\split_data.py", line 23, in split_data X = Xy.drop(columns=['Load']) TypeError: drop() got an unexpected keyword argument 'columns'

  Thanks for your help!
hatalis commented 5 years ago

'Load' was part of a data set on power systems, when using your own data for detection you will need to alter the code (eg name of the columns) with domain specifics of your data. Hope that helps!

ghost commented 5 years ago

I used your code and your data, and did not change anything. Could you try it by yourself? Thanks!

Tommylushixiang commented 4 years ago

Thank you for your contribution， I tried your code and got this error:

python detector.py line 52, in model.compile(loss=lambda Y, Q: pinball_loss(tau, Y, Q, alpha, smooth_loss, kappa, margin), optimizer=optimizer) ValueError: Tensor-typed variable initializers must either be wrapped in an init_scope or callable (e.g., tf.Variable(lambda : tf.truncated_normal([10, 40])))