Gijs van Tulder gvtulder

In this commit https://github.com/Theano/Theano/pull/5190/commits/772fbcb00ba03b8ef9c6c198f04bf519a91589b1 I 'fixed' a bug by renaming the property 'inplace_running_mean' to 'inplace_running_xxx'.

The problem seems to be in the optimization:

There is a GpuDnnBatchNorm Op that takes as its first input a tensor ('x'). This tensor is then normalized and returned as the output. The GpuDnnBatchNorm Op has a few inplace parameters named inplace_running_mean, inplace_running_var and inplace_output. Setting inplace_output=True will modify the original input tensor 'x'.
The gradient Op GpuDnnBatchNormGrad takes as input the original tensor ('x') and some outputs of the GpuDnnBatchNorm Op. This means that the grad Op has to run after the normalization Op.
Obviously, if the normalization Op and the grad Op use the same input 'x', the normalization Op should not run with inplace_output=True.

	Encrypting hard drives without reinstalling Ubuntu
	===================================================
	Gijs van Tulder, 26.02.2019, last update 04.03.2019

	This is a list of the steps I took to encrypt the partitions on my
	Ubuntu 16.04 laptop with the LUKS encryption system. Encrypting the
	partitions took some time but was relatively easy. Getting the system
	to boot afterwards was a little trickier, but doable.

	Obviously, these steps might not work for you. Follow them at your

	from __future__ import print_function
	import numpy

	import theano
	import theano.tensor as T
	from theano.tensor.nnet.abstract_conv import AbstractConv2d
	from theano.tensor.nnet.abstract_conv import AbstractConv2d_gradInputs
	from theano.tensor.nnet.corr import CorrMM_gradInputs
	from theano.gpuarray.blas import GpuCorrMM_gradInputs
	from theano.gpuarray.dnn import dnn_gradinput

	from __future__ import print_function

	import time
	from collections import OrderedDict

	import numpy as np

	import theano
	import theano.tensor as T
	import theano.gpuarray