Keras Constant Input Layers with Fixed Source of Stochasticity

Louis Tiao

2017-11-17 01:57

In [1]:

%matplotlib notebook

In [2]:

import numpy as np
import keras.backend as K

from keras.layers import Input, Activation, Add, GaussianNoise
from keras.models import Sequential, Model

Using TensorFlow backend.

Rationale¶

TODO

In [3]:

random_tensor = K.random_normal(shape=(8, 3), seed=42)

In [4]:

K.eval(random_tensor)

Out[4]:

array([[-0.28077507, -0.13775212, -0.67632961],
       [ 0.02458041, -0.89358455, -0.82847327],
       [ 1.2068944 ,  1.38101566, -1.45579767],
       [-0.24621388, -1.36084056,  1.08796036],
       [-0.35116589, -0.51385337,  3.41172075],
       [ 0.05885483,  0.89180237, -0.7528832 ],
       [-0.4335728 ,  2.45385313,  0.31374422],
       [-0.52736205,  0.85249925, -0.5379132 ]], dtype=float32)

In [5]:

K.eval(random_tensor)

Out[5]:

array([[ 0.36944136, -0.06497762,  1.05423534],
       [ 0.92629176,  0.45142221,  0.6538806 ],
       [ 0.00987345, -0.75727743,  1.19744813],
       [ 0.10721783, -1.34733653,  0.69856125],
       [ 1.34215105,  0.19264366, -0.02015864],
       [ 0.61278504,  0.43748191,  1.21581125],
       [ 0.42827308, -1.2276696 , -2.39826727],
       [-0.21679108,  0.05826041,  0.10147382]], dtype=float32)

In [6]:

x = Input(shape=(784,))
eps = Input(tensor=K.random_normal(shape=(K.shape(x)[0], 3), seed=42))

In [7]:

try:
    m = Model(x, eps)
except RuntimeError as e:
    print(e)

Graph disconnected: cannot obtain value for tensor Tensor("random_normal_1:0", shape=(?, 3), dtype=float32) at layer "input_2". The following previous layers were accessed without issue: []

In [8]:

m = Model([x, eps], eps)

In [9]:

m.predict(np.ones((8, 784)))

Out[9]:

array([[-0.28077507, -0.13775212, -0.67632961],
       [ 0.02458041, -0.89358455, -0.82847327],
       [ 1.2068944 ,  1.38101566, -1.45579767],
       [-0.24621388, -1.36084056,  1.08796036],
       [-0.35116589, -0.51385337,  3.41172075],
       [ 0.05885483,  0.89180237, -0.7528832 ],
       [-0.4335728 ,  2.45385313,  0.31374422],
       [-0.52736205,  0.85249925, -0.5379132 ]], dtype=float32)

In [10]:

m.predict([np.ones((8, 784))])

Out[10]:

array([[ 0.36944136, -0.06497762,  1.05423534],
       [ 0.92629176,  0.45142221,  0.6538806 ],
       [ 0.00987345, -0.75727743,  1.19744813],
       [ 0.10721783, -1.34733653,  0.69856125],
       [ 1.34215105,  0.19264366, -0.02015864],
       [ 0.61278504,  0.43748191,  1.21581125],
       [ 0.42827308, -1.2276696 , -2.39826727],
       [-0.21679108,  0.05826041,  0.10147382]], dtype=float32)

In [11]:

try:
    m.predict([np.ones((8, 784)), np.ones((8, 3))])
except ValueError as e:
    print(e)

Error when checking model : the list of Numpy arrays that you are passing to your model is not the size the model expected. Expected to see 1 array(s), but instead got the following list of 2 arrays: [array([[ 1.,  1.,  1., ...,  1.,  1.,  1.],
       [ 1.,  1.,  1., ...,  1.,  1.,  1.],
       [ 1.,  1.,  1., ...,  1.,  1.,  1.],
       ..., 
       [ 1.,  1.,  1., ...,  1.,  1.,  1.],
       [ 1...