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.

dlib/all/source.cpp into their own .cpp files.  If so they will get a useful
error message.

visual studio.

sys.stdout.encoding instead of latin-1 in setup.py

Add pragma guards around deprecated auto_ptr to prevent GCC warnings

Fixes #67

Please use sys.stdout.encoding instead of latin-1 in setup.py.
This is necessary for non English OS.

network (which would needlessly double VRAM usage).  I also added a
set_synchronization_file() method so you can tell it to automatically
synchronize itself to disk every so often during training.  This makes resuming
an interrupted training session trivially easy.

the loss isn't being reduced.  Also, there is a stopping condition now based on
how large the current learning rate is.  That is, training stops when the learning
rate gets small enough and it is clear that no progress is being made.