tjwei / quick-neural-art-transfer Goto Github PK
View Code? Open in Web Editor NEWTheano/Lasagne based Neural artistic style transfer with Kivy GUI
License: MIT License
Theano/Lasagne based Neural artistic style transfer with Kivy GUI
License: MIT License
The readme docs lack the proper procedure to run the scripts. Please update
The main.py when run throws an Attribute error. This is because class objects are being initialized using data of three different folders out of which "~./Output/ " is missing.
I get the following error from the prompt.
C:\Users\user\Documents\GitHub\Quick-Neural-Art-Transfer [master +0 ~1 -0]>
C:\Users\user\Documents\GitHub\Quick-Neural-Art-Transfer [master +0 ~1 -0]> python main.py
[INFO ] [Logger ] Record log in C:\Users\user\.kivy\logs\kivy_16-03-13_7.txt
[INFO ] [Kivy ] v1.9.1
[INFO ] [Python ] v2.7.10 (default, May 23 2015, 09:44:00) [MSC v.1500 64 bit (AMD64)]
[INFO ] [Factory ] 179 symbols loaded
[INFO ] [Image ] Providers: img_tex, img_dds, img_gif, img_sdl2, img_pil (img_ffpyplayer ignored)
[INFO ] [OSC ] using <thread> for socket
[INFO ] [Window ] Provider: sdl2
[INFO ] [GL ] GLEW initialization succeeded
[INFO ] [GL ] OpenGL version <4.2.0 - Build 10.18.10.3379>
[INFO ] [GL ] OpenGL vendor <Intel>
[INFO ] [GL ] OpenGL renderer <Intel(R) HD Graphics 4400>
[INFO ] [GL ] OpenGL parsed version: 4, 2
[INFO ] [GL ] Shading version <4.20 - Build 10.18.10.3379>
[INFO ] [GL ] Texture max size <16384>
[INFO ] [GL ] Texture max units <32>
[INFO ] [Window ] auto add sdl2 input provider
[INFO ] [Window ] virtual keyboard not allowed, single mode, not docked
content/tjw2.jpg
1227
(1227L, 1443L, 3L)
(dtype('uint8'), dtype('float64'))
output/tjw_a.png
Traceback (most recent call last):
File "main.py", line 73, in <module>
("content/smile_face.jpg", "styles/gh.jpg", "output/c5.png")]]
File "main.py", line 55, in __init__
self.output = get_img(output)
File "main.py", line 47, in get_img
print len(img)
TypeError: object of type 'NoneType' has no len()
C:\Users\user\Documents\GitHub\Quick-Neural-Art-Transfer [master +0 ~1 -0]>
I cannot start nserver.py it reports:
(myenv) ubuntu@ip-172-31-15-252:~/Quick-Neural-Art-Transfer$ python nserver.py Using gpu device 0: GRID K520 (CNMeM is enabled with initial size: 75.0% of memory, cuDNN Version is too old. Update to v5, was 3007.) /home/ubuntu/myenv/lib/python3.4/site-packages/theano/tensor/signal/downsample.py:6: UserWarning: downsample module has been moved to the theano.tensor.signal.pool module. "downsample module has been moved to the theano.tensor.signal.pool module.") unable to import Conv2DDNNLayer, use Conv2DLayer instead Traceback (most recent call last): File "nserver.py", line 3, in <module> import neuralStyle File "/home/ubuntu/Quick-Neural-Art-Transfer/neuralStyle.py", line 69, in <module> net = build_model() File "/home/ubuntu/Quick-Neural-Art-Transfer/neuralStyle.py", line 44, in build_model net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1) File "/home/ubuntu/Quick-Neural-Art-Transfer/neuralStyle.py", line 35, in ConvLayer return ConvLayer0(*args, flip_filters=False, **kwargs) File "/home/ubuntu/myenv/lib/python3.4/site-packages/lasagne/layers/conv.py", line 391, in __init__ super(Conv2DLayer, self).__init__(incoming, **kwargs) TypeError: __init__() got an unexpected keyword argument 'flip_filters'
Any ideas? I'm using the latest version from git. I already fixed the type in super(Conv2DLayer, self).__init__(incoming, **kw**e**args)
to super(Conv2DLayer, self).__init__(incoming, **kwargs)
Upgrading to cuDNN v5 does not help.
Any help welcome.
hi @tjwei
I followed the instructions and finished the installation .
Then I run “python nstest.py” and get an error:
Traceback (most recent call last):
File "nstest.py", line 2, in
import neuralStyle
File "/home/sojoyoo/neural_style_kivy/neuralStyle.py", line 69, in
net = build_model()
File "/home/sojoyoo/neural_style_kivy/neuralStyle.py", line 44, in build_model
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
File "/home/sojoyoo/neural_style_kivy/neuralStyle.py", line 35, in ConvLayer
return ConvLayer0(*args, flip_filters=False, **kweargs)
NameError: global name 'kweargs' is not defined
Do you know how to solve this ?
Thank you
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.