twak / bikegan Goto Github PK
View Code? Open in Web Editor NEWfrankenGAN networks
License: Other
frankenGAN networks
License: Other
I'm trying to run your code with a 512x512 paired dataset (like in pix2pix models) so the train images are 1024x512, but every time I try
python train.py --dataroot ./datasets/08K_Rplan_Black --name 08K_RPlan_Black --model bicycle_gan --use_dropout --loadSize 512 --fineSize 512 --display_winsize 512
I GOT:
File "C:\Users\Pichau\anaconda3\envs\py37\lib\site-packages\torch\nn\functional.py", line 1370, in linear
ret = torch.addmm(bias, input, weight.t())
RuntimeError: size mismatch, m1: [1 x 1024], m2: [256 x 8] at C:/w/1/s/windows/pytorch/aten/src\THC/generic/THCTensorMathBlas.cu:290
Even when I specify other --which_model_netE resnet512
like junyanz sugested on the original source code, but then the code freezes and doesn't process any image.
Could you tell what networks were used to process results with the dataset 512x512 you used in the article... it seems this repo is the only one BicycleGAN repository that implemented networks that support bigger images in comparison to the 256x256...
NOTE: i've used the sabe dataset to train a model using the original source code, but it requires imgs to be 256* and the results are not satisfatory.
HELP PLZ <3
Hello, I ran "python test_interactive.py" and the result is shown below. Does this mean that pre-trained model weights have been downloaded? Also, I want to test the model. Could you provide some test data?
Loading model bicycle_gan...
---------- Networks initialized -------------
initialization method [xavier]
/mnt/FrankenGAN/bikegan/models/networks.py:27: UserWarning: nn.init.xavier_norma l is now deprecated in favor of nn.init.xavier_normal_.
init.xavier_normal(m.weight.data, gain=1.0)
initialization method [xavier]
/mnt/FrankenGAN/bikegan/models/networks.py:29: UserWarning: nn.init.xavier_norma l is now deprecated in favor of nn.init.xavier_normal_.
init.xavier_normal(m.weight.data, gain=1.0)
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network labels2door_5 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network r3_clabels2labels_f001_400 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network r3_labels2image_f001_400 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network w3_empty2labels_f009_200 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network w3_labels2image_f013_400 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network empty2windows_f009v2_400 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network facade_windows_f013v2_150 is awaiting input]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network image2celabels_f001_335 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
---------- Networks initialized -------------
[network super6 is awaiting input]
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
model [BiCycleGANModel] was created
all nets up
[network super10 is awaiting input]
Looking forward to your reply!
Could you provide the dataset for training or testing BikeGan?
I'm getting into a strange permission issue that could very well be a problem with my setup, but I'm sort of out of ideas on what to do.
I'm trying to run BikeGAN using nvidia-docker2
. The installation went perfectly, no problems with running basic tests with it either. Now, with working directory, [home]/bikegan-temp
, if I try to run:
> docker run --runtime=nvidia -v $(pwd)/input:/home/user/bikegan/input -v $(pwd)/output:/home/user/bikegan/output -it --rm twak/bikegan
That gives the following error:
docker: Error response from daemon: error while creating mount source path '[home]/bikegan-temp/input': mkdir [home]/bikegan-temp/input: permission denied.
Okay, what if I try to mkdir input
and mkdir output
myself to [home]/bikegan-temp
, and try to rerun the command?
> docker run --runtime=nvidia -v $(pwd)/input:/home/user/bikegan/input -v $(pwd)/output:/home/user/bikegan/output -it --rm twak/bikegan
Loading model bicycle_gan...
---------- Networks initialized -------------
initialization method [xavier]
/home/user/bikegan/models/networks.py:27: UserWarning: nn.init.xavier_normal is now deprecated in favor of nn.init.xavier_normal_.
init.xavier_normal(m.weight.data, gain=1.0)
initialization method [xavier]
/home/user/bikegan/models/networks.py:29: UserWarning: nn.init.xavier_normal is now deprecated in favor of nn.init.xavier_normal_.
init.xavier_normal(m.weight.data, gain=1.0)
-----------------------------------------------
model [BiCycleGANModel] was created
Loading model bicycle_gan...
Unhandled exception in thread started by <bound method Interactive.go of <__main__.Interactive object at 0x7f386059feb8>>
Traceback (most recent call last):
File "/home/user/bikegan/test_interactive.py", line 331, in go
os.makedirs(input_folder + "val", exist_ok=True)
File "/home/user/miniconda/envs/py36/lib/python3.6/os.py", line 210, in makedirs
makedirs(head, mode, exist_ok)
File "/home/user/miniconda/envs/py36/lib/python3.6/os.py", line 220, in makedirs
mkdir(name, mode)
PermissionError: [Errno 13] Permission denied: './input/door textures'
---------- Networks initialized -------------
initialization method [xavier]
initialization method [xavier]
-----------------------------------------------
This error prints out ~10 times, each time with a different thing that has permission denied:
PermissionError: [Errno 13] Permission denied: './input/roof greebles'
PermissionError: [Errno 13] Permission denied: './input/roof textures'
PermissionError: [Errno 13] Permission denied: './input/pane labels'
PermissionError: [Errno 13] Permission denied: './input/pane textures'
PermissionError: [Errno 13] Permission denied: './input/facade labels'
PermissionError: [Errno 13] Permission denied: './input/facade textures'
PermissionError: [Errno 13] Permission denied: './input/facade greebles'
PermissionError: [Errno 13] Permission denied: './input/facade super'
PermissionError: [Errno 13] Permission denied: './input/roof super'
My system:
Ubuntu 14.04.5 LTS
I've tried to install and uninstall docker, checked that my system doesn't use Docker Snap, and tried most things that the internet has told me. Any other ideas?
In "models/base_model.py", line 235
cuda(, async=True) --> cuda(, non_blocking=True)
because cuda( ) replaced async using non_blocking argument.
I'm running python train.py --dataroot ./datasets/08K_Rplan_Black --name 08K_RPlan_Black --model bicycle_gan --dataset_mode aligned --which_model_netE resnet_512 --use_dropout --loadSize 512 --fineSize 512 --display_winsize 512
on a python 3.6 env
nvcc --version returns:
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Oct_12_20:54:10_Pacific_Daylight_Time_2020
Cuda compilation tools, release 11.1, V11.1.105
Build cuda_11.1.relgpu_drvr455TC455_06.29190527_0
and
print(torch.version.cuda)
print(torch. cuda. is_available())
print(torch. cuda. get_device_name(torch. cuda. current_device()))
returns:
10.1
True
GeForce RTX 2060
Error logs:
CustomDatasetDataLoader
dataset [AlignedDataset] was created
#training images = 800
Loading model bicycle_gan...
---------- Networks initialized -------------
Traceback (most recent call last):
File "train.py", line 26, in <module>
model = create_model(opt)
File "C:\Users\Pichau\bikegan-master\models\__init__.py", line 10, in create_model
model.initialize(opt)
File "C:\Users\Pichau\bikegan-master\models\bicycle_gan_model.py", line 32, in initialize
self.init_data(opt, use_D=use_D, use_D2=use_D2, use_E=use_E, use_vae=True)
File "C:\Users\Pichau\bikegan-master\models\base_model.py", line 46, in init_data
gpu_ids=self.gpu_ids, where_add=self.opt.where_add, upsample=opt.upsample)
File "C:\Users\Pichau\bikegan-master\models\networks.py", line 128, in define_G
assert(torch.cuda.is_available())
AssertionError
What torch version and cuda version should I use to be able to run the code...
a requirements.txt for pip would be useful
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.