In the Convolution2D docs keras , I found that there are no details regarding the definitionpadding
keras.layers.convolutional.Convolution2D(nb_filter,
nb_row,
nb_col,
init='glorot_uniform',
activation=None,
weights=None,
border_mode='valid',
subsample=(1, 1),
dim_ordering='default',
W_regularizer=None,
b_regularizer=None,
activity_regularizer=None,
W_constraint=None,
b_constraint=None,
bias=True)
argument subsample
tuple of length 2. Factor with which you can select the output. Also called steps elsewhere.
and I think this is a step.
And argument border_mode
'valid', 'same' or 'full'. ("full" requires the Theano backend.)
validand sameare also arguments to the conv2d TensorFlow function.
How to determine the filling, how to set its value?
source
share