[Bug] Error with "Positional Encoding as Spatial Inductive Bias in GANs" inference result have artifact

[junseokoh1]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• I have read the FAQ documentation but cannot get the expected help.
• The bug has not been fixed in the latest version.

Task

I'm using the official example scripts/configs for the officially supported tasks/models/datasets.

Environment

The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
If you want to use the GeForce RTX 3090 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

warnings.warn(incompatible_device_warn.format(device_name, capability, " ".join(arch_list), device_name))
sys.platform: linux
Python: 3.8.0 (default, Nov 6 2019, 21:49:08) [GCC 7.3.0]
CUDA available: True
CUDA_HOME: /usr/local/cuda
NVCC: Build cuda_11.0_bu.TC445_37.28540450_0
GPU 0: GeForce RTX 3090
GPU 1: GeForce GTX 1080 Ti
GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
PyTorch: 1.8.0
PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
• Intel(R) MKL-DNN v1.7.0 (Git Hash 7aed236906b1f7a05c0917e5257a1af05e9ff683)
• OpenMP 201511 (a.k.a. OpenMP 4.5)
• NNPACK is enabled
• CPU capability usage: AVX2
• CUDA Runtime 10.2
• NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70
• CuDNN 7.6.5
• Magma 2.5.2
• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=10.2, CUDNN_VERSION=7.6.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -O2 -fPIC
  -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.8.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

TorchVision: 0.9.0
OpenCV: 4.7.0
MMCV: 1.7.0
MMGen: 0.7.2+f9f00fd
MMCV Compiler: GCC 7.3
MMCV CUDA Compiler: 10.2

Reproduces the problem - code sample

I used official code from your github.
Nothing chaged.

Reproduces the problem - command or script

Comment that i used

For fish
CUDA_VISIBLE_DEVICES=1 python3 tools/utils/singan_inference.py configs/positional_encoding_in_gans/singan_csg_fish.py
/home/uhrgan/Neurocomputing/mmgeneration/infer_param/singan_csg_fis_20210406_175532-f0ec7b61.pth --samples-path output/fish --save-prev-res --seed 22

For bohemina - CSG
CUDA_VISIBLE_DEVICES=1 python3 tools/utils/singan_inference.py configs/positional_encoding_in_gans/singan_csg_bohemian.py
/home/uhrgan/Neurocomputing/mmgeneration/infer_param/singan_csg_bohemian_20210407_195455-5ed56db2.pth --samples-path output/bohemina_csg --save-prev-res --seed 22

For bohemina - SPE
CUDA_VISIBLE_DEVICES=1 python3 tools/utils/singan_inference.py configs/positional_encoding_in_gans/singan_spe-dim4_bohemian.py
/home/uhrgan/Neurocomputing/mmgeneration/infer_param/singan_spe-dim4_bohemian_20210406_175820-6e484a35.pth --samples-path output/bohemina_spe --save-prev-res --seed 22

Following guide line I also add test_cfg to sigan_bohemian.py & singan_fish.py file.

config file that i used.

base = [
'../base/models/singan/singan.py', '../base/datasets/singan.py',
'../base/default_runtime.py'
]

num_scales = 10 # start from zero
model = dict(
generator=dict(num_scales=num_scales),
discriminator=dict(num_scales=num_scales))

train_cfg = dict(
noise_weight_init=0.1,
iters_per_scale=2000,
)

test_cfg = dict(
delete = True,
pkl_data = '/home/uhrgan/Neurocomputing/mmgeneration/infer_param/singan_csg_fis_20210406_175532-f0ec7b61.pkl'
)

data = dict(
train=dict(
img_path='./data/singan/fish-crop.jpg', min_size=25, max_size=300))

optimizer = None
lr_config = None
checkpoint_config = dict(by_epoch=False, interval=2000, max_keep_ckpts=3)

custom_hooks = [
dict(
type='MMGenVisualizationHook',
output_dir='visual',
interval=500,
bgr2rgb=True,
res_name_list=['fake_imgs', 'recon_imgs', 'real_imgs']),
dict(
type='PickleDataHook',
output_dir='pickle',
interval=-1,
after_run=True,
data_name_list=['noise_weights', 'fixed_noises', 'curr_stage'])
]

total_iters = 22000

But result have artifact

Reproduces the problem - error message

There is no error message, but result have artifacts

Additional information

No response

[junseokoh1]

Results for bohemina. Also there is few artifact, and result for SPE have changed color