/Python/When training= false still dropout

When training= false still dropout

from keras import layers
from keras import models
from keras import backend as K
import numpy as np

inp = layers.Input(shape=(10,))
out = layers.Dropout(0.5)(inp)

model = models.Model(inp, out)
model.layers[-1].trainable = False  # set dropout layer as non-trainable
model.compile(optimizer='adam', loss='mse') # IMPORTANT: we must always compile model after changing `trainable` attribute

# create a custom backend function so that we can control the learning phase
func = K.function(model.inputs + [K.learning_phase()], model.outputs)

x = np.ones((1,10))
# learning phase = 1, i.e. training mode
print(func([x, 1]))
# the output will be:
[array([[2., 2., 2., 0., 0., 2., 2., 2., 0., 0.]], dtype=float32)]
# as you can see some of the neurons have been dropped

# now set learning phase = 0, i.e test mode
print(func([x, 0]))
# the output will be:
[array([[1., 1., 1., 1., 1., 1., 1., 1., 1., 1.]], dtype=float32)]
# unsurprisingly, no neurons have been dropped in test phase

Source: stackoverflow.com

Similar codes

-2 votes178 views1 answer
Requests.exceptions.connectionerror: ('connection aborted.', remotedisconnected('remote end closed connection without response'))
Python
0 vote194 views1 answer
Schedule task di windows 7 dengan php
Python
20 votes186 views6 answers
Python generate random password
Python
0 vote167 views1 answer
Python generate sine wave pyaudio
Python
4 votes178 views3 answers
Fibonacci sequence python code
Python
1 vote159 views1 answer
Colab save figure
Python
66 votes165 views6 answers
How to open csv file in python
Python
0 vote160 views2 answers
Fibonacci series python program
Python
0 vote175 views1 answer
IPython default setup
Python
0 vote165 views3 answers
Write a Python program to count total number of notes in given amount.
Python