Daiki Sanno

Thank you for your comment. In "Selecting S^t and S^s" section of original paper, the authors mention two methods to rank images. - counting thenumber of matching attributes (e.g., matching...

@mazzzystar > But is that similar with the VGG feature distance ? Yes, it is. > I worried these feature conficts with the final VGG model and can lead to...

Are there any console messages? If the console is showing "load VGG16 caffemodel", the program should be loading caffe model file. It takes several minutes, and please be patient.

Can you create CaffeFunction instance? Please confirm if CaffeFunction constructor works as follows: ``` (on Python interactive shell) >>> from chainer.links.caffe import CaffeFunction >>> ref = CaffeFunction('VGG_ILSVRC_16_layers.caffemodel') ``` If it...

The correct path for VGG_ILSVRC_16_layers.caffemodel is "chainer-neural-style". In more detail 1. Put VGG_ILSVRC_16_layers.caffemodel into chainer-neural-style directory. 2. Move to chainer-neural-style directory 3. Run Python and create CaffeFunction instance. You can...

Did you check VGG_ILSVRC_16_layers.caffemodel exists **on Python shell** ? For example, do these commands work? ``` >>> import os >>> os.getcwd() '/path/to/current/working/directory' >>> os.path.exists('VGG_ILSVRC_16_layers.caffemodel') True >>> f = open('VGG_ILSVRC_16_layers.caffemodel', 'rb')...

@jmic00 Thank you for your report `cuda.Device` depends on 'cupy' module doesn't work without 'cupy', so I shouldn't have use it. I fixed to use `cuda.get_device_from_id` instead of it.

It is usual to consume 2.5G memory. VGG model takes a large amount of memory. Memory usage depends on image size and you can reduce memory usage by using `-w...

Thank you for your reply. Chainer supports Adam optimizer and it is easy to use it. To use Adam, replace this line(https://github.com/dsanno/chainer-neural-style/blob/master/src/style_transfer_runner.py#L52) with `optimizer = chainer.optimizers.Adam(10.0)` (I don't know the...

> 1、Do you know how to limit the gpu usage(for example limit it in 1G) in this chainer neural style code. I think Chainer doesn't have a function to limit...