R-package/demo/basic_random.R - mxnet - Git at Google

 require(mxnet)

 mx.set.seed(10)

 print(mx.runif(c(2,2), -10, 10))

 # Test initialization module for neural nets.
 uinit <- mx.init.uniform(0.1)
 print(uinit("fc1_weight", c(2, 2), mx.cpu()))
 print(uinit("fc1_gamma", c(2, 2), mx.cpu()))
	require(mxnet)

	mx.set.seed(10)

	print(mx.runif(c(2,2), -10, 10))

	# Test initialization module for neural nets.
	uinit <- mx.init.uniform(0.1)
	print(uinit("fc1_weight", c(2, 2), mx.cpu()))
	print(uinit("fc1_gamma", c(2, 2), mx.cpu()))