szha commented on issue #7408: How to do a sobel operation in symbol layer URL: https://github.com/apache/incubator-mxnet/issues/7408#issuecomment-321460822 Something like this? The example is in ndarray but you can pass the kernel values in through `sym.var`. ``` from mxnet import ndarray as nd data = nd.random_uniform(shape=(1, 1, 300, 300)) kx = nd.array([[1, 2, 1]]).T * nd.array([[1, 0, -1]]) ky = kx.T.reshape((1, 1, 3, 3)) kx = kx.reshape((1, 1, 3, 3)) gx = nd.Convolution(data, kx, kernel=(3, 3), num_filter=1, no_bias=True, layout='NCHW') gy = nd.Convolution(data, ky, kernel=(3, 3), num_filter=1, no_bias=True, layout='NCHW') out = nd.sqrt(gx*gx + gy*gy) ``` ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
With regards, Apache Git Services