doesn't have it.

representation rather than dlib's default variable length encoding.  This
change makes the resulting serialized networks about 33% smaller.

objects.  Then made the relevant parts of the code use these functions.

often create a conflict with MATLAB's copy of the MKL.

after training has finished.

entire std::vector of objects.

unit tests for the routines supporting this feature.

implementation of assign_conv_bias_gradient().

wrong outputs sometimes.

Decode message only if encoding is known

When python does not know the encoding of stdout, sys.stdout.encoding
is None.  Then calling decode(None) raises an exception.  We just
skip decoding when the encoding is unknown.

running_gradient object.

if you try to open a jpeg or png file.