top_level = pyramid[-1]
content_image = top_level.resize_image(content_image)
style_image = top_level.resize_image(style_image)
########################################################################################
# As a last preliminary step the previously loaded images are set as targets for the
# perceptual loss (``criterion``) and we create the input image.
criterion.set_content_image(content_image)
criterion.set_style_image(style_image)
starting_point = "content"
input_image = get_input_image(starting_point, content_image=content_image)
show_image(input_image, title="Input image")
########################################################################################
# Finally we run the NST with the
# :func:`~pystiche.optim.optim.default_image_pyramid_optim_loop`. If ``get_optimizer``
# is not specified, as is the case here, the
# :func:`~pystiche.optim.optim.default_image_optimizer`, i.e.
# :class:`~torch.optim.lbfgs.LBFGS` is used.
output_image = default_image_pyramid_optim_loop(input_image,
criterion,
pyramid,
logger=demo_logger())
# sphinx_gallery_thumbnail_number = 4
show_image(output_image, title="Output image")
# The optimization runs on each ``level`` for ``level.num_steps``.
#
#
# In every step perceptual loss is calculated
# with the ``criterion`` and propagated backward to the ``input_image``. If
# ``get_optimizer`` is not specified, as is the case here, the
# :func:`~pystiche.optim.optim.default_image_optimizer`, i.e.
# :class:`~torch.optim.lbfgs.LBFGS` is used.
#
# .. note::
#
# By default ``pystiche`` logs the time during an optimization. In order to reduce
# the clutter, we use the minimal ``demo_logger`` here.
output_image = default_image_optim_loop(
input_image, criterion, num_steps=500, logger=demo_logger()
)
########################################################################################
# After the NST is complete we show the result.
# sphinx_gallery_thumbnail_number = 4
show_image(output_image, title="Output image")
########################################################################################
# Conclusion
# ----------
#
# If you started with the basic NST example without ``pystiche`` this example hopefully
def test_demo_logger_smoke(caplog):
optim_logger = demo.demo_logger()
with asserts.assert_logs(caplog, logger=optim_logger):
optim_logger.message("test message")
def test_demo_logger_smoke(self):
optim_logger = demo.demo_logger()
with self.assertLogs(optim_logger.logger, "INFO"):
optim_logger.message("Test message")