AERΒΆ

path: orion.primitives.aer.AER

description: this an autoencoder-based model capable of creating both prediction-based and reconstruction-based anomaly scores.

see json.

argument

type

description

parameters

X

numpy.ndarray

n-dimensional array containing the input sequences for the model

y

numpy.ndarray

n-dimensional array containing the target sequences we want to reconstruct. Typically y is a signal from a selected set of channels from X.

hyperparameters

epochs

int

number of epochs to train the model. An epoch is an iteration over the entire X data provided

input_shape

tuple

tuple denoting the shape of an input sample

optimizer

str

string (name of optimizer) or optimizer instance. Default is keras.optimizers.Adam

learning_rate

float

float denoting the learning rate of the optimizer. Default is 0.001

batch_size

int

number of samples per gradient update. Default is 64

layers_encoder

list

list containing layers of encoder

layers_generator

list

list containing layers of generator

output

ry_hat

numpy.ndarray

n-dimensional array containing the regression for each input sequence (reverse)

y_hat

numpy.ndarray

n-dimensional array containing the reconstructions for each input sequence

fy_hat

numpy.ndarray

n-dimensional array containing the regression for each input sequence (forward)

In [1]: import numpy as np

In [2]: from mlstars import load_primitive

In [3]: X = np.ones((64, 100, 1))

In [4]: y = X[:,:, [0]] # signal to reconstruct from X (channel 0)

In [5]: primitive = load_primitive('orion.primitives.aer.AER',
   ...:     arguments={"X": X, "y": y, "epochs": 1, "batch_size": 1})
   ...: 

In [6]: primitive.fit()

 1/51 [..............................] - ETA: 3:05 - loss: 0.6328 - tf.__operators__.getitem_loss: 0.7061 - tf.__operators__.getitem_1_loss: 0.5557 - tf.__operators__.getitem_2_loss: 0.7139
 4/51 [=>............................] - ETA: 1s - loss: 0.3926 - tf.__operators__.getitem_loss: 0.5086 - tf.__operators__.getitem_1_loss: 0.2902 - tf.__operators__.getitem_2_loss: 0.4815  
 7/51 [===>..........................] - ETA: 0s - loss: 0.2571 - tf.__operators__.getitem_loss: 0.3485 - tf.__operators__.getitem_1_loss: 0.1846 - tf.__operators__.getitem_2_loss: 0.3108
10/51 [====>.........................] - ETA: 0s - loss: 0.2191 - tf.__operators__.getitem_loss: 0.2498 - tf.__operators__.getitem_1_loss: 0.2042 - tf.__operators__.getitem_2_loss: 0.2180
13/51 [======>.......................] - ETA: 0s - loss: 0.1913 - tf.__operators__.getitem_loss: 0.1988 - tf.__operators__.getitem_1_loss: 0.1983 - tf.__operators__.getitem_2_loss: 0.1698
16/51 [========>.....................] - ETA: 0s - loss: 0.1660 - tf.__operators__.getitem_loss: 0.1762 - tf.__operators__.getitem_1_loss: 0.1702 - tf.__operators__.getitem_2_loss: 0.1476
19/51 [==========>...................] - ETA: 0s - loss: 0.1495 - tf.__operators__.getitem_loss: 0.1685 - tf.__operators__.getitem_1_loss: 0.1445 - tf.__operators__.getitem_2_loss: 0.1405
22/51 [===========>..................] - ETA: 0s - loss: 0.1386 - tf.__operators__.getitem_loss: 0.1652 - tf.__operators__.getitem_1_loss: 0.1253 - tf.__operators__.getitem_2_loss: 0.1386
25/51 [=============>................] - ETA: 0s - loss: 0.1296 - tf.__operators__.getitem_loss: 0.1608 - tf.__operators__.getitem_1_loss: 0.1108 - tf.__operators__.getitem_2_loss: 0.1361
28/51 [===============>..............] - ETA: 0s - loss: 0.1214 - tf.__operators__.getitem_loss: 0.1537 - tf.__operators__.getitem_1_loss: 0.1003 - tf.__operators__.getitem_2_loss: 0.1313
31/51 [=================>............] - ETA: 0s - loss: 0.1143 - tf.__operators__.getitem_loss: 0.1448 - tf.__operators__.getitem_1_loss: 0.0938 - tf.__operators__.getitem_2_loss: 0.1248
34/51 [===================>..........] - ETA: 0s - loss: 0.1087 - tf.__operators__.getitem_loss: 0.1359 - tf.__operators__.getitem_1_loss: 0.0904 - tf.__operators__.getitem_2_loss: 0.1179
37/51 [====================>.........] - ETA: 0s - loss: 0.1040 - tf.__operators__.getitem_loss: 0.1281 - tf.__operators__.getitem_1_loss: 0.0879 - tf.__operators__.getitem_2_loss: 0.1120
40/51 [======================>.......] - ETA: 0s - loss: 0.0997 - tf.__operators__.getitem_loss: 0.1220 - tf.__operators__.getitem_1_loss: 0.0847 - tf.__operators__.getitem_2_loss: 0.1074
43/51 [========================>.....] - ETA: 0s - loss: 0.0959 - tf.__operators__.getitem_loss: 0.1177 - tf.__operators__.getitem_1_loss: 0.0807 - tf.__operators__.getitem_2_loss: 0.1043
46/51 [==========================>...] - ETA: 0s - loss: 0.0925 - tf.__operators__.getitem_loss: 0.1146 - tf.__operators__.getitem_1_loss: 0.0766 - tf.__operators__.getitem_2_loss: 0.1022
49/51 [===========================>..] - ETA: 0s - loss: 0.0896 - tf.__operators__.getitem_loss: 0.1121 - tf.__operators__.getitem_1_loss: 0.0729 - tf.__operators__.getitem_2_loss: 0.1005
51/51 [==============================] - 6s 49ms/step - loss: 0.0877 - tf.__operators__.getitem_loss: 0.1104 - tf.__operators__.getitem_1_loss: 0.0707 - tf.__operators__.getitem_2_loss: 0.0992 - val_loss: 0.0420 - val_tf.__operators__.getitem_loss: 0.0647 - val_tf.__operators__.getitem_1_loss: 0.0191 - val_tf.__operators__.getitem_2_loss: 0.0651

In [7]: ry, y, fy = primitive.produce(X=X)

1/2 [==============>...............] - ETA: 0s
2/2 [==============================] - 0s 10ms/step

1/2 [==============>...............] - ETA: 0s
2/2 [==============================] - 0s 11ms/step

In [8]: print("Reverse Prediction: {}\nReconstructed Values: {}, Forward Prediction: {}".format(ry, y, fy))
Reverse Prediction: [[0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]
 [0.74567188]]
Reconstructed Values: [[[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]

 [[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]

 [[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]

 ...

 [[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]

 [[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]

 [[0.86373213]
  [0.94708711]
  [1.00604448]
  ...
  [0.98981457]
  [0.93391949]
  [0.85568408]]], Forward Prediction: [[0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]
 [0.74485171]]