QGAC issueshttps://gitlab.com/Queuecumber/quantization-guided-ac/-/issues2023-02-11T17:21:38Zhttps://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/12Train the model with own dataset2023-02-11T17:21:38ZYuanming CaoTrain the model with own datasetHi,
Thanks for such amazing project, I'm wondering is it possible of releasing training code for other datasets?
Best regardsHi,
Thanks for such amazing project, I'm wondering is it possible of releasing training code for other datasets?
Best regardshttps://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/11Output result2022-02-20T07:29:34ZMariam AliOutput resultHello @Queuecumber, I was wondering if there is something other than the provided command (whether for a single image or batch) that I have to edit/add to try the algorithm on my image(s); 'cause when I have tried the output is almost th...Hello @Queuecumber, I was wondering if there is something other than the provided command (whether for a single image or batch) that I have to edit/add to try the algorithm on my image(s); 'cause when I have tried the output is almost the same as the input (and sometimes even worse)
Ps. I have tried the provided weights and merp between themhttps://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/10How can high resolution images be handled2022-02-16T14:50:02ZMariam AliHow can high resolution images be handledHello @Queuecumber,
I really appreciate your work, but when I have tested it on high resolution images (ex. 1092x1080) it gave cuda out of memory on an 8GB GPU. <br>
`CUDA out of memory. Tried to allocate 1.45 GiB (GPU 0; 8.00 GiB total...Hello @Queuecumber,
I really appreciate your work, but when I have tested it on high resolution images (ex. 1092x1080) it gave cuda out of memory on an 8GB GPU. <br>
`CUDA out of memory. Tried to allocate 1.45 GiB (GPU 0; 8.00 GiB total capacity; 5.35 GiB already allocated; 425.25 MiB free; 5.60 GiB reserved in total by PyTorch)`<br>
So, Is there a tweak or alternative other than downsizing the input image.
Thanks in advancehttps://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/9Torchjpeg installation issue2021-10-23T21:18:00ZShashank VasishtTorchjpeg installation issueI am unable to install torchjpeg. It exits with status code 1
```
(qgac) C:\Users\shashank\Desktop\Fynd\Codes\quantization-guided-ac>pip install torchjpeg==0.9.14
Collecting torchjpeg==0.9.14
Using cached torchjpeg-0.9.14.tar.gz (1.0...I am unable to install torchjpeg. It exits with status code 1
```
(qgac) C:\Users\shashank\Desktop\Fynd\Codes\quantization-guided-ac>pip install torchjpeg==0.9.14
Collecting torchjpeg==0.9.14
Using cached torchjpeg-0.9.14.tar.gz (1.0 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing wheel metadata ... done
Requirement already satisfied: torchvision==0.8.2 in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torchjpeg==0.9.14) (0.8.2)
Requirement already satisfied: torch==1.7.1 in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torchjpeg==0.9.14) (1.7.1)
Requirement already satisfied: Pillow>=7.2.0 in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torchjpeg==0.9.14) (8.3.2)
Requirement already satisfied: typing_extensions in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torch==1.7.1->torchjpeg==0.9.14) (3.10.0.2)
Requirement already satisfied: numpy in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torch==1.7.1->torchjpeg==0.9.14) (1.19.5)
Requirement already satisfied: dataclasses in c:\users\shashank\anaconda3\envs\qgac\lib\site-packages (from torch==1.7.1->torchjpeg==0.9.14) (0.8)
Building wheels for collected packages: torchjpeg
Building wheel for torchjpeg (PEP 517) ... error
ERROR: Command errored out with exit status 1:
command: 'c:\users\shashank\anaconda3\envs\qgac\python.exe' 'c:\users\shashank\anaconda3\envs\qgac\lib\site-packages\pip\_vendor\pep517\in_process\_in_process.py' build_wheel 'C:\Users\shashank\AppData\Local\Temp\tmpq7jf5agj'
cwd: C:\Users\shashank\AppData\Local\Temp\pip-install-4r8pkju1\torchjpeg_7b4cc014cf2440a0b5f27215bc1028bb
Complete output (25 lines):
A setup.py file already exists. Using it.
Traceback (most recent call last):
File "C:\Users\shashank\AppData\Local\Temp\pip-install-4r8pkju1\torchjpeg_7b4cc014cf2440a0b5f27215bc1028bb\setup.py", line 2, in <module>
from setuptools import setup
ModuleNotFoundError: No module named 'setuptools'
Traceback (most recent call last):
File "c:\users\shashank\anaconda3\envs\qgac\lib\site-packages\pip\_vendor\pep517\in_process\_in_process.py", line 349, in <module>
main()
File "c:\users\shashank\anaconda3\envs\qgac\lib\site-packages\pip\_vendor\pep517\in_process\_in_process.py", line 331, in main
json_out['return_val'] = hook(**hook_input['kwargs'])
File "c:\users\shashank\anaconda3\envs\qgac\lib\site-packages\pip\_vendor\pep517\in_process\_in_process.py", line 249, in build_wheel
metadata_directory)
File "C:\Users\shashank\AppData\Local\Temp\pip-build-env-e46sxlm2\overlay\Lib\site-packages\poetry\core\masonry\api.py", line 68, in build_wheel
return unicode(WheelBuilder.make_in(poetry, Path(wheel_directory)))
File "C:\Users\shashank\AppData\Local\Temp\pip-build-env-e46sxlm2\overlay\Lib\site-packages\poetry\core\masonry\builders\wheel.py", line 72, in make_in
wb.build()
File "C:\Users\shashank\AppData\Local\Temp\pip-build-env-e46sxlm2\overlay\Lib\site-packages\poetry\core\masonry\builders\wheel.py", line 103, in build
self._build(zip_file)
File "C:\Users\shashank\AppData\Local\Temp\pip-build-env-e46sxlm2\overlay\Lib\site-packages\poetry\core\masonry\builders\wheel.py", line 135, in _build
self._run_build_command(setup)
File "C:\Users\shashank\AppData\Local\Temp\pip-build-env-e46sxlm2\overlay\Lib\site-packages\poetry\core\masonry\builders\wheel.py", line 169, in _run_build_command
str(self._path / "build"),
File "c:\users\shashank\anaconda3\envs\qgac\lib\subprocess.py", line 311, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['c:/users/shashank/anaconda3/envs/qgac/python.exe', 'C:\\Users\\shashank\\AppData\\Local\\Temp\\pip-install-4r8pkju1\\torchjpeg_7b4cc014cf2440a0b5f27215bc1028bb\\setup.py', 'build', '-b', 'C:\\Users\\shashank\\AppData\\Local\\Temp\\pip-install-4r8pkju1\\torchjpeg_7b4cc014cf2440a0b5f27215bc1028bb\\build']' returned non-zero exit status 1.
----------------------------------------
ERROR: Failed building wheel for torchjpeg
Failed to build torchjpeg
ERROR: Could not build wheels for torchjpeg which use PEP 517 and cannot be installed directly
```https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/8pickle.UnpicklingError: invalid load key, 'v'2021-06-02T23:08:18ZPetteri Teikaripickle.UnpicklingError: invalid load key, 'v'Hi,
Thanks for sharing such an interesting repo, I tried to test the approach for my data, and ran into the following invalid code when reading some DCT stats (?)
Any idea what is going on?
```
python experiments/correct_batch.py --in...Hi,
Thanks for sharing such an interesting repo, I tried to test the approach for my data, and ran into the following invalid code when reading some DCT stats (?)
Any idea what is going on?
```
python experiments/correct_batch.py --input $INPUT_DIR test --output $OUTPUT_DIR --weights ./weights/weights_c.pt --dstats_path ./stats/cstats.pt --color
Traceback (most recent call last):
File "experiments/correct_batch.py", line 51, in <module>
stats = dct.DCTStats(args.dstats_path, type="ms")
File "quantization-guided-ac/dct/dct_stats.py", line 7, in __init__
super(DCTStats, self).__init__(root, type)
File "quantization-guided-ac/dct/stats.py", line 7, in __init__
self.stats = torch.load(root)
File "/home/petteri/anaconda3/lib/python3.8/site-packages/torch/serialization.py", line 593, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/home/petteri/anaconda3/lib/python3.8/site-packages/torch/serialization.py", line 762, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, 'v'.
```https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/7RuntimeError: Unable to open file for reading: experiments_on_python/x.jpg2021-04-11T04:40:15ZARSHIYA KHANRuntimeError: Unable to open file for reading: experiments_on_python/x.jpgTraceback (most recent call last):
File "/content/drive/MyDrive/Year-4/quantization-guided-ac-master/experiments/correct_single.py", line 64, in <module>
dim, quantization, y, cbcr = torchjpeg.codec.read_coefficients(args.input)
`R...Traceback (most recent call last):
File "/content/drive/MyDrive/Year-4/quantization-guided-ac-master/experiments/correct_single.py", line 64, in <module>
dim, quantization, y, cbcr = torchjpeg.codec.read_coefficients(args.input)
`RuntimeError: Unable to open file for reading: experiments_on_python/x.jpg`https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/6Sizes of tensors must match except in dimension 2. Got 60 and 120 (The offend...2021-01-14T17:57:09ZVinayak SharmaSizes of tensors must match except in dimension 2. Got 60 and 120 (The offending index is 0)Getting a shape mismatch for directories despite the same exact image characteristics across all images. It also seems sporadic as other folders with similar creation methods do not have this issue. Any help identifying the cause would b...Getting a shape mismatch for directories despite the same exact image characteristics across all images. It also seems sporadic as other folders with similar creation methods do not have this issue. Any help identifying the cause would be helpful.
* I have also confirmed that every single image in the folder works when using correct_batch.
* I think this issue was raised in this [Issue](https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/3), however the delta is much larger than they observed
``` bash
Traceback (most recent call last):
File "experiments/correct_batch.py", line 62, in <module>
y_r, cb_r, cr_r = net(y_q, y, c_q, cb, cr)
File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/data1/vinayak/quantization-guided-ac/model/deartifacting_network.py", line 88, in forward
r_cb = self.color_net(q_y, q_c, restored, cb)
File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/data1/vinayak/quantization-guided-ac/model/color_restore.py", line 42, in forward
d = self.block_enhancer2(torch.cat([d, y], dim=1))
RuntimeError: Sizes of tensors must match except in dimension 2. Got 60 and 120 (The offending index is 0)
root@7743ed28fae9:/data1/vinayak/quantization-guided-ac#
```
P.S Great work on the code though. Its well structured and mostly easy to use and I found your paper very interesting.https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/3sizes of tensors don't match when running the testing code2021-01-07T07:57:41ZJiaxi Jiangsizes of tensors don't match when running the testing codeHi Max,
When I tried to test the code on the LIVE1 dataset, the following error occurs. Testing on other datasets would have a similar error; Only a few images can generate a result. Could you help me with this problem? Thank you!
> py...Hi Max,
When I tried to test the code on the LIVE1 dataset, the following error occurs. Testing on other datasets would have a similar error; Only a few images can generate a result. Could you help me with this problem? Thank you!
> python experiments/correct_batch.py --input testsets/LIVE1_Color/10/ --output results/LIVE1_Color/10/ --weights ./weights/weights_c.pt --dstats_path ./stats/cstats.pt --color
> Correcting: 10% 3/29 [00:03<00:29, 1.12s/batch]
> Traceback (most recent call last):
> File "experiments/correct_batch.py", line 60, in <module>
> y_r, cb_r, cr_r = net(y_q, y, c_q, cb, cr)
> File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 722, in _call_impl
> result = self.forward(*input, **kwargs)
> File "./model/deartifacting_network.py", line 88, in forward
> r_cb = self.color_net(q_y, q_c, restored, cb)
> File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 722, in _call_impl
> result = self.forward(*input, **kwargs)
> File "./model/color_restore.py", line 42, in forward
> d = self.block_enhancer2(torch.cat([d, y], dim=1))
> RuntimeError: Sizes of tensors must match except in dimension 2. Got 61 and 62 (The offending index is 0)https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/5About the results with ARCNN in LIVE12021-01-05T14:17:40ZJuZiSYJAbout the results with ARCNN in LIVE1Hi!
I noticed that the numerical results in your paper about the PSNR and SSIM are largely **lower** than the shown in other papers, such as ARCNN in LIVE1-QF-10. The result shown in the other papers is (28.96/0.808/28.77), while that in...Hi!
I noticed that the numerical results in your paper about the PSNR and SSIM are largely **lower** than the shown in other papers, such as ARCNN in LIVE1-QF-10. The result shown in the other papers is (28.96/0.808/28.77), while that in your paper is (26.66/0.792/26.54).https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/4Import error, no module found.2020-12-24T00:12:09ZQLaHPDImport error, no module found.I keep getting this error:
```
Traceback (most recent call last):
File "../experiments/correct_single.py", line 3, in <module>
import dct
ModuleNotFoundError: No module named 'dct'
```
I tried to move all files to same folder, but...I keep getting this error:
```
Traceback (most recent call last):
File "../experiments/correct_single.py", line 3, in <module>
import dct
ModuleNotFoundError: No module named 'dct'
```
I tried to move all files to same folder, but looks like in order to do that I need to change the code a little, so how to solve this problem?https://gitlab.com/Queuecumber/quantization-guided-ac/-/issues/2ImportError and undefined symbol2020-09-23T18:53:31ZJiaxi JiangImportError and undefined symbolHello,
when I try to predict my own jpeg files using the provided pretrained model, the following error occurs:
Traceback (most recent call last):
File "experiments/correct_batch.py", line 8, in <module>
import dataset
File "./...Hello,
when I try to predict my own jpeg files using the provided pretrained model, the following error occurs:
Traceback (most recent call last):
File "experiments/correct_batch.py", line 8, in <module>
import dataset
File "./dataset/__init__.py", line 1, in <module>
from .jpeg_quantized_dataset import *
File "./dataset/jpeg_quantized_dataset.py", line 2, in <module>
import torchjpeg.codec
File "/conda_envs/cv/lib/python3.7/site-packages/torchjpeg/codec/__init__.py", line 14, in <module>
from ._codec_ops import *
ImportError: /conda_envs/cv/lib/python3.7/site-packages/torchjpeg/codec/_codec_ops.cpython-37m-x86_64-li
nux-gnu.so: undefined symbol: _ZN6caffe28TypeMeta21_typeMetaDataInstanceISt7complexIfEEEPKNS_6detail12TypeMetaDataEv
my package version:
torch 1.6.0
torchvision 0.7.0