quolc / neural-collage Goto Github PK
View Code? Open in Web Editor NEWCollaging on Internal Representations: An Intuitive Approach for Semantic Transfiguration
License: MIT License
Collaging on Internal Representations: An Intuitive Approach for Semantic Transfiguration
License: MIT License
Hi!
First, type your code
python3 demo_spatial_translation.py --config ./configs/sn_projection_dog_and_cat_256_scbn.yml --gen_model ./sn_projection_dog_and_cat_256/ResNetGenerator_450000.npz --gpu 0
Then, I use GPU clusters, so that I only can open browser use command line.
For example,
user@user:opera
,
Then X11 will open.
But how to load image in that page?
Thanks.
argument = "--config_path ./configs/sn_projection_dog_and_cat_256_auxab.yml \
--gpu 0 \
--gen_model ./sn_projection_dog_and_cat_256_large/ResNetAuxABGenerator_75000.npz \
--dis_model ./sn_projection_dog_and_cat_256_large/SNResNetProjectionDiscriminator_450000.npz \
--enc_model ./sn_projection_dog_and_cat_256_large/ResNetEncoder_450000.npz \
--src_class {} \
--input patch_src.png \
--iter_opt 200 \
--mode aux".format(src_class)
This line can't run because zsh: number expected
on oh-my-zsh and bash: 未预期的符号
(' 附近有语法错误` on bash
python 3.5
File "demo_feature_blending.py", line 6, in
import secrets
ImportError: No module named 'secrets'
你好,我想请问是否可以使用你的代码对celeba人脸数据集进行训练?因为我看datasets里有imagenet和dog_and_cat的.py文件
Thank you for sharing.
In your paper you said you explored to enable the semantic modification of real images in "real time". So how fast did you get for optimization in your environment? In my try, it seems to be difficult to achieve real-time speed with 1 RTX 2080.
And did you try projection via a feedforward network?
Thanks.
Hello, I run NeuralCollage_demo.ipynb in colab. It show TypeError. How can it work pretty?
# perform blending
arguments = "--gpu 0 \
--config ./configs/sn_projection_dog_and_cat_256_scbn.yml \
--snapshot sn_projection_dog_and_cat_256_large/ResNetGenerator_450000.npz \
--results_dir log/log/gen_spatially_interpolated_images_with_feature_blend \
--z1 ./src.npy \
--z2 ./ref.npy \
--class_mask ./examples/mask_1.png \
--classes {} {}".format(target_class, src_class)
!python evaluations/gen_spatially_interpolated_images_with_feature_blend.py $arguments
Output
(45, 99)
Traceback (most recent call last):
File "evaluations/gen_spatially_interpolated_images_with_feature_blend.py", line 141, in <module>
main()
File "evaluations/gen_spatially_interpolated_images_with_feature_blend.py", line 124, in main
x = gen.spatial_interpolation(zs=[z1, z2], weights=ws, blends=blends)
File "gen_models/resnet_256_scbn.py", line 77, in spatial_interpolation
hs.append(chainer.Variable(self.xp.zeros_like(hs[0]))) # dummy
File "/usr/local/lib/python3.7/dist-packages/cupy/creation/basic.py", line 238, in zeros_like
shape)
File "/usr/local/lib/python3.7/dist-packages/cupy/creation/basic.py", line 42, in _new_like_order_and_strides
order = chr(_update_order_char(a, ord(order)))
TypeError: Argument 'x' has incorrect type (expected cupy.core.core.ndarray, got Variable)
RuntimeError Traceback (most recent call last)
in ()
1 import chainer
----> 2 chainer.cuda.get_device(0).use()
/usr/local/lib/python3.6/dist-packages/chainer/backends/cuda.py in get_device(*args)
224 warnings.warn('get_device is deprecated. Please use get_device_from_id or'
225 ' get_device_from_array instead.', DeprecationWarning)
--> 226 return _get_device(*args)
227
228
/usr/local/lib/python3.6/dist-packages/chainer/backends/cuda.py in _get_device(*args)
230 for arg in args:
231 if type(arg) is not bool and isinstance(arg, _integer_types):
--> 232 check_cuda_available()
233 return Device(arg)
234 if isinstance(arg, ndarray):
/usr/local/lib/python3.6/dist-packages/chainer/backends/cuda.py in check_cuda_available()
91 '(see https://github.com/chainer/chainer#installation).')
92 msg += str(_resolution_error)
---> 93 raise RuntimeError(msg)
94 if (not cudnn_enabled and
95 not _cudnn_disabled_by_user and
RuntimeError: CUDA environment is not correctly set up
(see https://github.com/chainer/chainer#installation).CuPy is not correctly installed.
If you are using wheel distribution (cupy-cudaXX), make sure that the version of CuPy you installed matches with the version of CUDA on your host.
Also, confirm that only one CuPy package is installed:
$ pip freeze
If you are building CuPy from source, please check your environment, uninstall CuPy and reinstall it with:
$ pip install cupy --no-cache-dir -vvvv
Check the Installation Guide for details:
https://docs-cupy.chainer.org/en/latest/install.html
original error: libcublas.so.9.2: cannot open shared object file: No such file or directory
get this error in the google colab notebook
Hi~ Thank you for your excellent work! I am wondering if you could also upload a demo for human face images as well. I'd really appreciate that!
Thank you for greate paper!
I have a question about a manifold projection method for StyleGAN generator.
Did you use the same way where the loss function is described in Equation (4) in paper to obtain the corresponding latent/style representation of a given image?
Thanks in advance!
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.