km2ii / jvp_tg-plus Goto Github PK
View Code? Open in Web Editor NEWThis is enhanced mod for javplayer, enhance the effect after removing mosaic.
This is enhanced mod for javplayer, enhance the effect after removing mosaic.
I have cuda 11.0 installed and I can't install cuda 11.1 due to incompatibility with another application, it would have some way of working on 11.0, I did the installation, but I see little use of dedicated video card and Vram 0.2 in cuda 11.0
Hi,
I keep getting this error regarding time.log. It seems to be present on longer recordings more often. Ive tried reinstalling everything twice but I still get this error
Some mosaic areas have deformation or jitter. How to set and solve it?
Firstly I'm running this with an AMD Ryzen 5950X and Nvidia GTX 1080ti. I have all packs installed and using the following simple mod.ini
TG-PLUS
# This is a JP109A_TG-PLUS profile.
# Note the first line cannot be removed.
M1:tecogan:4:TG-STD
M2:tecogan:4:SE-4X
V1.9
RES:1920
When I run this processing on a file using Safe mode I have observed only my CPU being utilised and my GPU untouched. And yes I have selected my GPU in the advanced tecogan settings.
Due to this, processing takes at least 5 times longer compared to the unmodded version of JAVPlayer which does utilise the GPU Cuda cores. A 15 second 1080p clip takes over 10 minutes to process on TG-PLUS compared to less than 2 minutes on the unmodded version. This makes it almost unusable for longer movies.
Is this to be expected?
I installed the new TG-PLUS v.1.08.09 on a fresh copy of JavPlayer.
After the Extraction phase JavPlayer displays the error:
Failed to convert images
No images have been converted at all
I resolved it by manually creating my own config.ini, so it seems it cannot create the config file on the first run.
PS. Thank you for your great work on this.
Any Planned work to add the Dione models for veai? Running TG-AF1 to reduce flicker makes things blurry, I find in other use that the Dione models do the best to bring details back.
Do you have any idea if the mosaic can not detect from old movie ?
i upscale it from 480P to 720P first , will little bit better , but some of mosaic still cant detect ....
Hi , i Just check GPU Utilization is 0% and CPU is 70% when SE-4X is working , i have set the javplayer using GPU0 by TG and FFmpeg , the performance is it right or anything wrong ?
may km2ii write the denoise parameter name or introduction in readme, thx a lot
denoise almost works, now hyb throws an error [info] Denoising please wait....
Script evaluation failed:
Python exception: No module named 'muvsfunc'
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1946, in vapoursynth.vpy_evaluateScript
File "src\cython\vapoursynth.pyx", line 1947, in vapoursynth.vpy_evaluateScript
File "Vapoursynth\hyb.vpy", line 7, in
ModuleNotFoundError: No module named 'muvsfunc'
Script evaluation failed:
Python exception: No module named 'muvsfunc'
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1946, in vapoursynth.vpy_evaluateScript
File "src\cython\vapoursynth.pyx", line 1947, in vapoursynth.vpy_evaluateScript
File "Vapoursynth\hyb.vpy", line 7, in
ModuleNotFoundError: No module named 'muvsfunc'
Script evaluation failed:
Python exception: No module named 'muvsfunc'
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1946, in vapoursynth.vpy_evaluateScript
File "src\cython\vapoursynth.pyx", line 1947, in vapoursynth.vpy_evaluateScript
File "Vapoursynth\hyb.vpy", line 7, in
ModuleNotFoundError: No module named 'muvsfunc'
TG-PLUS_v.1.08.17.rar (499.75 MB)
Fixed some bugs, no new features, no need to re-download.
https://workupload.com/file/NYvUanSKLT6
this one is down
why not use mega or googledrive ?
tgmain.exe is flagged as gen:variant.bulz.566620 by my antivirus
by bitdefender
Not just me, several guys from fourm also claimed these plugs don't work for them.
For me, it's 3060 ,cuda11.1, cudnn 8.05, when Javplayer works, there is a 100% cpu usage for most part of process.
I copied the dlls to TGMAIN, not working, I think if there is a discord group will be help us reporting this errors.
I generated several videos with your plugs( using cpu), the TG+VAE result is fabulous! I can't use original javplayer back again. XD
Although the results are far better than the original javplayer, but the gpu adaptation seems need to be fix.
Thanks for your working!
Why the mod's TG-MODEL is smaller than TecoGAN 1.08 ?Can you update TG-MODEL ?
it works well for a shot part
but if select a long part, the mosaic area does not match the source frame
looks like their timeline doesn't match...
I just installed Veai 2.3.0, and using TG-PLUS with patch2.
When I using veai in mplus, it shows the error dialog below:
After checking Config.ini, I found that the "veaidict" section was cleared (this happens every time starting tgmain.exe)
Before:
After:
Is that a bug? Or there's something wrong in my configuration process?
Whenever I use TG Plus on older videos (480p) I get terrible results in comparison to HD videos and devensoring older videos with standard Javplayer as well. I've tried your example config files too (probably made for 720+). Any suggestion for config file settings to get better results for 480p video when compared to standard Javplayer?
sorry ,
if i train tecogan , change the dataset with differnent scenes ,should i retrain or continue train is ok?
sorry for asking this question , but i cant find others could help me
hey I see TG-plus for the first time and I would like to install it but I don't know how. I have read README_OLD.md but the links there do not work. can you help me there
Thank you for sharing your wonderful work. It seems like the download links are not working properly. Would you check it? Or it possible to provide Google Drive link?
Thank you.
Error Type <class 'AttributeError'>
'NoneType' object has no attribute 'astype'
<FrameSummary file corem.py. line 228 in corem>
<FrameSummary file utis.py. line 326 in utis.inference_data_loader>
<FrameSummary file utis.py, line 318 in
utils.inference_data_loader.preprocess_test>
how can i fix it?
EDIT JAN 1 2021: this is a universal TG-plus time limit to prevent people from maliciously selling this mod.
Original post:
Hi km2ii - javskii just released v1.10 for the base app yesterday, and I got the current TG-Plus (v0.1.10.24) to work flawlessly with it for the whole day (processed multiple vids without a hitch), but then it started giving the error "ERROR - Please update your TG-Plus!" whenever I processed something.
Since it worked flawlessly for a while, is there anything minor I can change for it to keep the compatibility? Or wait for an official new TG-plus release from you? Thanks!
when going to the m2 denosie:tla step , the cmd show
py_initialize unable to load the file system code no module named encodings
when i fix this problem , show other problem ,
fatal python error: py_initialize: can't initialize sys standard streams " io.py "
TG step and SE-4X-R1 no problem , just only the part of Denosie ,
how can i fix this ?
File "", line 665, in _load_unlocked
File "", line 929, in exec_module
File "", line 219, in _call_with_frames_removed
File "tf_slim\ops\arg_scope.py", line 184, in func_with_args
File "tf_slim\layers\layers.py", line 1449, in convolution2d_transpose
File "tensorflow\python\keras\engine\base_layer_v1.py", line 1696, in apply
File "tensorflow\python\keras\legacy_tf_layers\base.py", line 561, in call
File "tensorflow\python\keras\engine\base_layer_v1.py", line 783, in call
File "tensorflow\python\autograph\impl\api.py", line 692, in wrapper
File "tensorflow\python\autograph\impl\api.py", line 336, in converted_call
File "tensorflow\python\autograph\impl\api.py", line 463, in _call_unconverted
File "tensorflow\python\keras\layers\convolutional.py", line 1322, in call
File "tensorflow\python\util\dispatch.py", line 206, in wrapper
File "tensorflow\python\keras\backend.py", line 5397, in conv2d_transpose
File "tensorflow\python\util\dispatch.py", line 206, in wrapper
File "tensorflow\python\ops\nn_ops.py", line 2494, in conv2d_transpose
File "tensorflow\python\util\dispatch.py", line 206, in wrapper
File "tensorflow\python\ops\nn_ops.py", line 2579, in conv2d_transpose_v2
File "tensorflow\python\ops\gen_nn_ops.py", line 1291, in conv2d_backprop_input
File "tensorflow\python\framework\op_def_library.py", line 750, in _apply_op_helper
File "tensorflow\python\framework\ops.py", line 3565, in _create_op_internal
File "tensorflow\python\framework\ops.py", line 2045, in init
<FrameSummary file core_main.py, line 315 in core_main>
<FrameSummary file core_main.py, line 346 in core_main>
<FrameSummary file tensorflow\python\client\session.py, line 968 in run>
<FrameSummary file tensorflow\python\client\session.py, line 1191 in _run>
<FrameSummary file tensorflow\python\client\session.py, line 1369 in _do_run>
<FrameSummary file tensorflow\python\client\session.py, line 1394 in _do_call>
E:\JavPlayer_109a1>ffmpeg -r 23,86839 -n -i "E:/JavPlayer_109a1/TG/blend0/%04d.png" -c:v h264_nvenc -gpu 0 -pix_fmt yuv420p -b:v 7778k "E:/JavPlayer_109a1/Movie/#0.mp4"#0.mp4"
ffmpeg version git-2020-01-08-5bd0010 Copyright (c) 2000-2020 the FFmpeg developers
built with gcc 9.2.1 (GCC) 20191125
configuration: --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt --enable-amf
libavutil 56. 38.100 / 56. 38.100
libavcodec 58. 65.102 / 58. 65.102
libavformat 58. 35.101 / 58. 35.101
libavdevice 58. 9.103 / 58. 9.103
libavfilter 7. 70.101 / 7. 70.101
libswscale 5. 6.100 / 5. 6.100
libswresample 3. 6.100 / 3. 6.100
libpostproc 55. 6.100 / 55. 6.100
[image2 demuxer @ 0000027d9a10a900] Unable to parse option value "23,86839" as video rate
[image2 demuxer @ 0000027d9a10a900] Error setting option framerate to value 23,86839.
E:/JavPlayer_109a1/TG/blend0/%04d.png: Invalid argument
ffmpeg1 -y -r 23,86839 -f image2 -i E:/JavPlayer_109a1/TG/blend0/%04d.png -vcodec libx264 -crf 15 -vf movie=TG/logo.png[watermark];[in][watermark]overlay=main_w-overlay_w-30:30 -pix_fmt yuv420p "E:/JavPlayer_109a1/Movie/
v1.08.12 [info] Denoising please wait ....
"vspipe.exe" is not internal or external
command, executable program, or batch file.
"vspipe.exe" is not internal or external
command, executable program, or batch file.
"vspipe.exe" is not internal or external
command, executable program, or batch file.
Its great to see your new verisons, you are a big big inventor .... thank you so much ,,
for the new function can i have some questions ?
Fixed SE-4X-R1 model << whats different now ?
I Love the Gaia Mode , protues also good , but Gaia is more sharpen, what is your opinion ?
m4 = downscale , may i know what is this use for ? why need downscale ?
i have many 480P video, do you thank i VEAI upscale to 720P first , the quality of removing mosaic and details will better ? or same ? My test result is almost same ,,,,,
i still dont know how to re-Gen the config.ini file ........crying
thank you inventor ...
Just letting you know the download link for the ConfigTools is broken
Do you train new Tecogan model?or just use the already version model ,Thx
I saw your sample , really really amazing !!! thank you your innovation !!
and the sample you bring with setting of config.ini
Questions :
thank you !
thanks for your great job again , is it positable to add one more option for the quality between SE-4X and SE-4X-R1 ? because R1 is loss too much details ...
I occasionally try to use this but I've never gotten through a full clip. I've noticed a few times I ran out of disk space when I had ~100gb available. Using this leaves veai set for png output when veai is restarted outside op tg-plus so I'm fairly certain it's setting veai up for png. My config is set to use jpg so it seems to be ignoring that for veai
Sorry, my language is poor, roughly as follows
The error rate of video synthesis is very high
The most common is,GetThreadContext failed and Picture missing, etc
For longer videos, it is not only time-consuming
In case of failure, you need to start over
Hope to continue after correcting the wrong position
I'm not sure if this issue is due to the mod, but I'm having an issue where whenever my JavPlayer is opened it will keep looping through all my previous videos until it eventually crashes.
I've tried deleting the JavPlayer.udj but after this it instead just says "the file played back last time cannot be found" and keeps looping.
I've even tried installing a fresh unmodified copy of JavPlayer and the same thing happens. This leads me believe JavPlayer has some sort of configuration elsewhere on the computer that may be corrupted. Do you have any idea where this is?
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.