python - Stride with different height and width in Tensorflow -
i'm using tensorflow process 1d signals (shape being (?, 5000, 1, 12)) , perform stride operations. however, when try tf.nn.conv2d(input_variable, w, strides=[1, stride, 1, 1], padding='same'), valueerror:
/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/common_shapes.pyc in conv2d_shape(op) 207 if stride_r != stride_c: 208 # todo(shlens): add support this. --> 209 raise valueerror("current implementation supports equal length " 210 "strides in row , column dimensions.") 211 valueerror: current implementation supports equal length strides in row , column dimensions. is there known workaround limits of current implementation, allow me have horizontal stride only?
we're working on supporting strides different lengths.
Comments
Post a Comment