😁 Nih, dude.. gua turunkan cara install A.I GFPGAN nya untuk Windows, simak bae bae ya.. [Untuk yang versi simple] Online Demo: kzitem.info/news/bejne/2m2HnmhqcaeBd4Y Tutorial lainnya seputar A.I ada dibawah ini: 1. Real-ESRGAN (General Image) 👉 kzitem.info/news/bejne/uqOkqoSlrYuflm0 2. Real-ESRGAN (General Image) via Google Colab 👉 kzitem.info/news/bejne/rYuFs41jm3OTYI4 3. GFPGAN (Face Restoration) 👉 kzitem.info/news/bejne/w2Z7p2t6s4BzhG0 4. GFPGAN (Face Restoration) via Google Colab 👉 kzitem.info/news/bejne/yadmvWydophznGU 5. Online Demo (General / Face Restoration) 👉 kzitem.info/news/bejne/2m2HnmhqcaeBd4Y 🙌 Dan terimakasih udah support terus channel NIHIL Project hingga saat ini.. terimakasih juga yang udah support, rutin ngelike atau mungkin ngeshare, karena dengan banyaknya like kalian sudah membantu konten-konten di Channel NIHIL Project menjadi direkomendasikan oleh KZitem. Dan terimakasih sawerianya: • Lembah Thanks THR nya ma dude~ 🥳🍻 ‼ Kalo kalian punya artwork bisa tag IG @nihilproject ya, gua udah bikin menu sorotan khusus untuk lu pada dude, baik praktik sehabis nonton konten disini atau apapun boleh-boleh aja..oke
@prorate8533
2 жыл бұрын
bang pytorch nya pilih CPU bisa apa enggak
@febriank3041
2 жыл бұрын
Mantap brader ilmu yang sangat bermanfaat👍🏽
@NIHILProject
2 жыл бұрын
@@prorate8533 bisa, sambil baca-baca juga ya dude official githubnya, sesuaikan devicenya. Semoga berhasil.
@NIHILProject
2 жыл бұрын
@@febriank3041 Tengkiuu bradeeerr, sehat-sehat selalu yak👍
@verocreator
2 жыл бұрын
Yang awam puseng Apa itu anaconda prempt dll, dan bukanya dmn?? Gada stap by stap untuk yg awam😵
@MLuigi-tj1xl
2 жыл бұрын
Buset, konten yg gak direncanain upload aja berkualitas gini, mengherankan emang abang yg satu ini 👏🏻
@NIHILProject
2 жыл бұрын
wkwk jangan terlalu heran, itu semua cuma karena nongkrong di kamar mandi
@gustianfajar11
Жыл бұрын
editan simple tapi keren, penyampaian mudah dimengerti, tutorial mudah diikuti mantap bang
@ManWithNoNameUomoSenzaNome
2 жыл бұрын
Thanks a bunch! And Thanks for adding subtitles!
@falqijackson6463
2 жыл бұрын
Sungkem dulu ah.... biar berkah ilmunya
@NIHILProject
2 жыл бұрын
Hii...Aamiin tengkiuu 😁
@Mfyus
Жыл бұрын
Alhamdulillah dapet chanell ini, dari kmrn gagal mulu install @_@ Thanks ya gan, udah subscribe
@aryotrikarsono1297
2 жыл бұрын
akhirnya bisa juga. waktu jalanin script-nya ada pesan error nggak ada cuda gpu, pas dicheck ternyata berhasil. trims tutorialnya, sangat bermanfaat👍.
@NIHILProject
2 жыл бұрын
hahaha Mantap, prosesnya bisa dilihat di Task Manager saat mulai running script enhance, apakah GPU atau CPU yang persenannya lebih tinggi. kalo CPU biasanya lebih lambat proses enhancenya. Developnya lebih menyarankan maksimalnya menggunakan GPU / NVIDIA+CUDA bila punya
@aryotrikarsono1297
2 жыл бұрын
@@NIHILProject siip, trims sarannya 🙏 jadi penasaran liat prosesnya, masih pake laptop keluaran 2013 yg pake Nvidia GeForce gt 650m ini, kyknya blom ada cuda nya 😅
@gmdoni3000
2 жыл бұрын
Gan.. Ane sudah sampai proses test restorasi foto Tapi Driver ane bukan NVIDIA Tulisannya : RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver Berarti tak iso ya?
@alvindevaprasetya8813
Ай бұрын
Bang izin bertanya, saya sudah megikuti tutorial diatas, namun ada yang bermasalah ketika running script Bagiamana ya bang ? File "D:\AI\GFPGAN\GFPGAN\inference_gfpgan.py", line 2, in import cv2 ModuleNotFoundError: No module named 'cv2
@nooinx
2 жыл бұрын
Akan Selalu Angkat Topi buat DUDE DUDE Nihil Project! Thank you DUDE!
@NIHILProject
2 жыл бұрын
Wih.. lulus lulusan dong 😂 thank you dude~ 🍻
@deaira7128
2 жыл бұрын
sangat2 bermanfaat, terimakasih banyak atas ilmunya gan.
@EnggaMuhammad
5 ай бұрын
kok pas bagian inferencenya ga bisa ya bang tulisan paling terakhir di log output kaya gini ls: cannot access 'results/cmp': No such file or directory. kal aja ada yg sama, thank you guys
@Meinnhart
Жыл бұрын
Works like a charm Thank you
@mahmudjapar5430
5 ай бұрын
Salam Kak... Maaf, "Setup.Py" nya tidak ada di folder "GFPGAN", jadi Step 6 tidak bisa; mohon pencerahannya Kak ... 🙏
@Gamer61895
6 ай бұрын
ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor' problem please Solve
@Jean-vq4qv
4 ай бұрын
Já tentei fazer pelo google drive e também tentei esse mesmo método do vídeo, toda vez que insiro as fotos ele dá erro e não gera as imagens.. O que eu estou fazendo de errado? Na linha final do código ele apresenta essa mensagem: "ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor'"
@abdullahfaqih2467
2 жыл бұрын
Mantap. Semakin sukses nihil project
@NIHILProject
2 жыл бұрын
Thanks brader dude 🙌🏻
@su4vecoisanenhuma
Жыл бұрын
pria Anda banyak membantu saya! terima kasih langsung dari brazil!
@Sensuka
2 жыл бұрын
Wow it works, thank you! I love you :D
@yelevenven5677
7 ай бұрын
pas masukin line terakhirnya : python inference_gfpgan.py -i inputs/upload -o results -v 1.3 -s 4 --bg_upsampler realesrgan muncul : ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor' (gak ada folder result di GFPGAN)
@aryanjohnsharma
Жыл бұрын
Thanks For Such a Polished Video!!! Here Have a Cookie🍪
@ImamAdries-yq7yd
5 ай бұрын
Bang punyaku folder result NY gak keluar ya bang terus ada tulisan modul not found error .please di bang
@Rafaelgodoyebert
5 ай бұрын
I have the error 'ImportError: DLL load failed:' how fix?
@kumirikidesign6407
Жыл бұрын
Ijin tanya dude, apakah bisa pas download pytorch nya dilempar ke directory lain selain :C ? Soalnya pas udh selesai download [Installing collecting package] muncul error Could not install packages due to an OSError
@mcfcxx9849
6 ай бұрын
Does anyone have a fix for the "ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor'" error?
@svytta
2 ай бұрын
TRY THIS: import fileinput import sys def update_degradations_import(file_path): #Original line we want to replace: original_line = 'from torchvision.transforms.functional_tensor import rgb_to_grayscale ' #New line that we want to write: new_line = 'from torchvision.transforms.functional import rgb_to_grayscale ' #Open the file in read and write mode: with fileinput.FileInput(file_path, inplace=True, backup='.bak') as file: for line in file: # Replace the original line with the new line: if line == original_line: sys.stdout.write(new_line) else: sys.stdout.write(line) #Path to the downgrade file that we want to modify, in my case: file_path = '/usr/local/lib/python3.10/dist-packages/basicsr/data/degradations.py' #Call function to update import update_degradations_import(file_path) print(f"Updated import in {file_path}")
@bd2.o30
10 күн бұрын
Are you using a graphic card? model No please?
@fahrul7856
2 жыл бұрын
Seperti biasa, ad lelucon yg gk bisa ditebak 😁👍👍👍
@NIHILProject
2 жыл бұрын
wahahaha kang pantau lelucon 😂
@thebyren
Жыл бұрын
muchas gracias por el tutorial, de verdad me sirvio sigue adelante
@opadarwin6151
2 жыл бұрын
saat download pytorch error, pdhal sudah proses download lwat anaconda promp 2,1 gb (kurang 40an mb) ERROR: Exception: Traceback (most recent call last): File "C:\Users\user\anaconda3\lib\site-packages\pip\_vendor\urllib3 esponse.py", line 438, in _error_catcher yield 😭any solution dude?
@ZaydKS
5 ай бұрын
😍
@danutw88
7 ай бұрын
mas pas menjalankan command ada keterangan, from torchvision.transforms.functional_tensor import rgb_to_grayscale ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor' solusinya gimana ya?
@mbentarrayat8905
4 ай бұрын
sama nih error nya
@edihindrariswandi8926
2 жыл бұрын
Mantaaap...saya suka banget nih channel
@NIHILProject
2 жыл бұрын
Thank you dudee 🍻
@yoniwoker
Жыл бұрын
este es un tutorial excelente.
@anatolyalikin8740
Жыл бұрын
Dear person! I've been looking for this opportunity for so long. And I did it!!! Thanks a lot!!! You are a very good person! I'm shocked!! Thank you friend!!
@mbentarrayat8905
4 ай бұрын
Waduh kok results nya sama aja ya sm foto yg di inputs/upload gada perubahan gtu
@laqisalas3677
2 жыл бұрын
gan, di scrip terakhir ketika jalankan proses penjernihan kok gagal, apa perlu pakek nvidea?
@bd2.o30
6 ай бұрын
Sir, Let me know your PC configuration Full details please. I tried many time, but not work.
@MSdoSER
2 жыл бұрын
HALP! I just keep getting "\GFPGAN\inference_gfpgan.py", line 2, in import cv2, ModuleNotFoundError: No module named 'cv2'" Any hints?
@mr.sneuna
2 жыл бұрын
akhirnya muncul juga videonya
@NIHILProject
2 жыл бұрын
Kalo gak muncul nanti ditanyain mulu 👀
@joko5363
Жыл бұрын
dude, bikin tutorial cara upgrade pip nya,ke versi baru. GFPGAN bagus buat yang fokus ke muka, tapi kalo full body ga semua jadi HD, cuma muka doang yang HD
@taufikirawan9892
11 ай бұрын
maaf min kenapa GFPGAN nya tidak bisa di clone ya
@maxsolar2788
Жыл бұрын
Goog video. Many thanks)!!!!
@Morgan-vb5li
Жыл бұрын
I LOVE YOU!thank you!
@MrMundink
2 жыл бұрын
instalasi selesai. tapi setelah mau eksekusi gambar ada error ModuleNotFoundError: No module named 'basicsr.losses.losses'. please help me
@NIHILProject
2 жыл бұрын
Pas proses instalasi apakah semua 100% complete tanpa adanya pesan error/gagal??
@colibrian
Жыл бұрын
thank you !🙂
@NIHILProject
Жыл бұрын
of course😸
@seranganalambaka7354
5 ай бұрын
emmm.. install basicSR nya gagal,
@sarevthok6838
2 жыл бұрын
RuntimeError: CUDA out of memory. Tried to allocate 632.00 MiB (GPU 0; 4.00 GiB total capacity; 2.04 GiB already allocated; 16.20 MiB free; 2.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF bang itu kenapa ya?
@NIHILProject
2 жыл бұрын
Perhatikan besar resolusi fotonya ya. Atau bisa coba tambahkan kode --tile 800 (atau dibawah 800) dude, contoh: python inference_realesrgan.py -n RealESRGAN_x4plus -i upload --outscale 3.5 --half --face_enhance --tile 800
@oktavinataranibu8699
2 жыл бұрын
Sangat membantu🤘😎
@NIHILProject
2 жыл бұрын
Semoga bermanfaat🤠
@irwansetiawan7042
2 жыл бұрын
saya udah install, tapi hasil nya lebih ke perbaikan wajah aja ya? kalo dari kain dan semacamnya ga di perbaikin, masih pecah, beda sama online. emang seperti itu kah ?
@NIHILProject
2 жыл бұрын
Prosesnya berjalan menggunakan CPU atau GPU? banyak yang terjadi kalau kasus seperti itu penggunaannya bukan dari GPU, dan AI yang memperbaiki gambar umum selain wajah adalah Real-ESRGAN dan Real-ESRGAN itu hanya bisa diproses dengan GPU, kalau agan prosesnya menggunakan CPU, AI nya lebih memfokuskan untuk hanya memperbaiki kualitas wajah saja. Apakah pesan saat instalasi GFPGAN berjalan normal/tidak ada code error?
@jeffrywae4443
2 жыл бұрын
Keren oiii 😍
@NIHILProject
2 жыл бұрын
Tengkiu oyy 😁
@estebanalvarez4530
Жыл бұрын
Hello, greetings, I have followed the steps for the installation, but when processing an image the background is totally black, it only processes the face, eliminating the background. What solution can there be?
@NIHILProject
Жыл бұрын
The issue has been discussed in their official post, please check the following link and try to follow the instructions given by the community to solve the problem. Good luck. github.com/TencentARC/GFPGAN/issues/205
@realWilliamMay
2 жыл бұрын
Thank you!
@nomberfax
Жыл бұрын
Brilliant
@hendrisatriamod9232
2 жыл бұрын
Download error dude, apa karna koneksi internet kah? Error : exception : traceback. Mohon pencerahannya. (Semoga karna koneksi internet aja)
@NIHILProject
2 жыл бұрын
Next content gua bakal share tentang kelanjutan photo enhancer ini dude
@maxsolar2788
Жыл бұрын
What changes do I need to make if I want to process the full photo, not just the face?
@NIHILProject
Жыл бұрын
GFPGAN = FACE REAL-ESRGAN = BACKGROUND 6:29 In the script, there is a command for upsampler-realesrgan, so it should automatically help improve the quality of the background.
@maxsolar2788
Жыл бұрын
@@NIHILProject Many thanks for your work. Wish you creative rise!
@wildarts
2 жыл бұрын
Bang kalo sudah selesai mengikuti step di video, apakah nanti GFPGAN nya bisa dijalankan secara offline ?
@NIHILProject
2 жыл бұрын
Bisa dude, setelah selesai install akan jadi portable
@wildarts
2 жыл бұрын
Baik bang , makasih banyak bang udh dijawab, Semoga Sehat selalu banyak rezeki abang sama tim.
@wildarts
2 жыл бұрын
Bang , sudah saya ikuti step by step nya seperti divideo, tapi pas mau upload foto buriq, malah error, dan gak ada folder result. Solusinya bang. Hiksss
@hafidansori
2 жыл бұрын
bang punya ku muncul keterangan begini kenapa ya ? cv2.error: OpenCV(4.5.5) D:\a\opencv-python\opencv-python\opencv\modules\imgcodecs\src\loadsave.cpp:730: error: (-2:Unspecified error) could not find a writer for the specified extension in function 'cv::imwrite_'
@hafidansori
2 жыл бұрын
apa karena format yang aku upload formatnya RAW ya ?
@NIHILProject
2 жыл бұрын
spek apa? foto lainnya bisa? resolusi fotonya berapa besar? test jpeg dulu aja buat train machine learningnya
@s1s1-9elap-Qu
2 жыл бұрын
bg gimana yah, di CMD sy malah muncul tulisan 'git' is not recognized as an internal or external command, operable program or batch file minta masukan nya bg, makasih
@NIHILProject
2 жыл бұрын
Cek deskripsi ada catatan disitu, pastikan gunakan Anaconda Prompt, bukan cmd dari windows.
@s1s1-9elap-Qu
2 жыл бұрын
@@NIHILProject udah bg, ternyata harus instal git distributed dulu
@kimiyalayn3940
2 жыл бұрын
Thanks😊 Hey! How you get tabs in your file explorer ?
@NIHILProject
2 жыл бұрын
it's Clover, extension for Windows Explorer
@idsf-tl2hu
2 жыл бұрын
selalu terbaik dude
@F002boy
Жыл бұрын
thanks!
@NIHILProject
Жыл бұрын
Welcome!
@ferribudi
Жыл бұрын
terima kasih gan tutornya, cara hapus/uninstalnya gimana gan? semacam pip dan sebagainya?
@NIHILProject
Жыл бұрын
Cari saja folder anaconda di drive C. Biasanya di "C:\Users\[Nama PC]\anaconda3\ buka folder "lib" delete "site-packages" Atau uninstall anaconda via control panel.
@Zwackelmann173
2 жыл бұрын
Thank you very much :D
@stefbaumeister266
Жыл бұрын
I tried this with the run code you provided and it successfully worked for a jpg. I tried using a mp4 file or a small video but got this code. if np.max(img) > 256: # 16-bit image TypeError: '>' not supported between instances of 'NoneType' and 'int' any idea?
@sirinsilvet
2 жыл бұрын
'git' is not recognized as an internal or external command, operable program or batch file. Muncul Gitu Bang Pas Mau Clone File Githubnya.
@NIHILProject
2 жыл бұрын
Cek kolom deskripsi ya dude 😲
@loritakute
Жыл бұрын
probably you have to reinstall git from window
@MessageFromHome
2 жыл бұрын
Thank you so much!!!
@NIHILProject
2 жыл бұрын
Hope it's useful😁
@satriaeritama3
2 жыл бұрын
Kak saya masukan foto satu badan utuh whole body di folder uploadnya. Tapi hasilnya di folder result hanya bagian muka saya yang diperbaiki selebihnya bagian badan lain jadi hitam. Bagaimana perbaikinya ya? Soalnya saya coba di google colape normal aja hasilnya bisa utuh diperbaiki g ada blank hitam. Petunjuk..
@NIHILProject
2 жыл бұрын
Ada pesan errornya kah saat running script via anaconda? Pastikan pas instalasi semuanya berjalan sukses ya, dan driver juga up to date.
@satriaeritama3
2 жыл бұрын
@@NIHILProject iya awalnya ada eror merah 1
@satriaeritama3
2 жыл бұрын
@@NIHILProject terus bagaimana min?
@auranous
2 жыл бұрын
'pip3' is not recognized as an internal or external command, operable program or batch file. muncul gini pas proses install Pytorch
@NIHILProject
2 жыл бұрын
Sudah menginstall Python nya? Lalu apakah menjalankan scriptnya sudah menggunakan Anaconda Prompt? Mohon disimak baik-baik ya, pelan dan santuy aja oke, pastikan persiapannya sudah siap semua, baru jalankan script satu-persatu melalui Anaconda Prompt, bukan melalui Command Prompt dari windows.
@ramdanihidayat3348
2 жыл бұрын
Ya ini permasalahan di karenakan tidak menggunakan anaconda prompt. Aku dah sukses install, dan work. Thanks bung Ega.
@ramdanihidayat3348
2 жыл бұрын
Windows 11 lancar.
@78karni
2 жыл бұрын
Saya juga mengalami serupa padahal sudah mengggunakan anaconda
@hakiki1518
2 жыл бұрын
Om, saya baru tertarik buat ngulik tutor ini, bisa kasih bocoran minimal spek pc ?
@NIHILProject
2 жыл бұрын
Semua proses dimulai dari coret-coretan kasar di kertas, lalu membuat skenario/cerita untuk bernarasi (intro adalah proses paling lama dalam pembuatan konten), lalu membuat storyboard, lalu mengumpulkan bahan dan membuat elemen grafis (photoshop), menghubungi pihak yang berhubungan dengan assets file (untuk izin agar tidak copyright claim, baik foto, gambar / music). Lalu malam/dini hari melakukan recording (audio/video). Pengeditan dimulai After Effects untuk pembuatan Intro(episode), Intro(case), Titles, Extra-scene(preparation), Saweria-Scene. Lalu Adobe Audition & FL Studio untuk mastering & enhance kualitas suara(narasi). Next, masuk Premiere untuk mixing Audio, Video, SFX, BGM, dsb. Lalu membuat thumbnail untuk KZitem, dan Blog(serta posting tulisan baru+link-linknya) Terakhir Aegisub untuk membuat subtitle Indonesia & Inggris. Spek PC masih standar aja belum upgrade besar-besaran, Ryzen 5 1600x up to 4.0Ghz, Ram 16GB (2x8), VGA 1050Ti (kadang ada yang minjemin 1660 Ti / RTX 3050)
@hakiki1518
2 жыл бұрын
@@NIHILProject ruarr biasa... 👍 semoga chanelnya maju ! Terimakasih banyak ilmunya !
@aprilsenjaapril
2 жыл бұрын
next boleh tuh bikin tutorial upscale video pake Real-ESRGAN ^^
@NIHILProject
2 жыл бұрын
uwadaaaw 😱
@gustavovieira8043
Жыл бұрын
This only works with a NVIDIA GPU or does it work with AMD also?
@NIHILProject
Жыл бұрын
I've read articles about this for suggestions to operate flawlessly with NVIDIA GPUs, however I haven't tried on devices with AMD GPUs. Furthermore, AMD GPU Torch only functions on Linux if Windows can only install CPU. bit.ly/3r7uKC9 Personally, I rarely use GFPGAN because Real-ESRGAN AI can now do the same thing with an additional script "-face enhance" (which is also based on gfpgan) Or else, you can now use the AI on this page (demo online method): replicate.com/xinntao/gfpgan or replicate.com/xinntao/realesrgan. I also discussed it in this video: kzitem.info/news/bejne/2m2HnmhqcaeBd4Y
@dhikasr318
Ай бұрын
Bang saya gabisa buka catatan nya kenapa
@faisallens
Жыл бұрын
Semua file yg udh didonload install dulu semua ya bang?
@NIHILProject
Жыл бұрын
Iya, jangan lupa install "git" juga ya buat pembaca bahasa pemrograman python, c++ dsb, lupa dimasukin di video, biasanya dipake buat clone repo dari github, yang perintahnya git clone "url"
@pierruno
7 ай бұрын
Can you change the title so people know it's not English?
@muhammadumarfarruq9010
Жыл бұрын
Bang tutorial di blogs nya apakah sudah di hapus? Kok saya buka tida bisa
@NIHILProject
Жыл бұрын
Bisa kok, nihilproject99.blogspot.com/2022/05/how-to-install-gfpgan-face-restoration.html
@adamsyahputra9376
2 жыл бұрын
Pernah coba ngitung gak, dalam satu gambar proses upscalingnya berapa detik??
@NIHILProject
2 жыл бұрын
Tergantung gpu dan besar resolusi pixel fotonya, tapi yang jelas untuk proses foto enhancer dengan A.I, bagi gua ini udah termasuk cepat! 💨
@alvindevaprasetya8813
2 жыл бұрын
Saya download manual PyTorch pakai IDM karena sering gagal faktor tidak stabilnya jaringan, izin bertanya file whl-nya ditaruh dimana ya bang setelah saya download ini ? Nuhun
@NIHILProject
2 жыл бұрын
oh perkara jaringan ya? belom coba ane kalo yang manual, defaultnya sih yang pasti di folder instalasi penyimpanan anaconda (tiap pc kadang berbeda lokasi penyimpanannya, jadi silahkan cari sendiri file penyimpanan anaconda ente ya) nanti simpan pytorchnya di .\anaconda3\lib\site-packages
@alvindevaprasetya8813
2 жыл бұрын
@@NIHILProject file dipindah dalam folder site-packages ya ? makasih bang
@muhamadrizkymuhaemin6236
2 жыл бұрын
cara download pytorch pakai idm gimana bang?
@w8prouser
Жыл бұрын
É possível manter a transparência de algumas imagens png?
@spyro0014
2 жыл бұрын
💯💯
@HutajuluProductions
3 ай бұрын
ma s saya kok gagal ya (base) C:\Users\HPJULU>cd project ai The system cannot find the path specified. (base) C:\Users\HPJULU>cd C:\Project AI\GFPGAN (base) C:\Project AI\GFPGAN>python inference_gfpgan.py -i inputs/upload -o results -v 1.3 -s 4 --bg_upsampler realesrgan Traceback (most recent call last): File "C:\Project AI\GFPGAN\inference_gfpgan.py", line 6, in import torch File "C:\Users\HPJULU\AppData\Roaming\Python\Python311\site-packages\torch\__init__.py", line 141, in raise err OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\HPJULU\AppData\Roaming\Python\Python311\site-packages\torch\lib\shm.dll" or one of its dependencies. (base) C:\Project AI\GFPGAN>
@sukmaid4134
2 жыл бұрын
mantappppppp
@030rahmatkamaldafiq5
2 жыл бұрын
Bang bagaimana cara membersihkan RAM dan Disk nya apabila sudah tidak muat, agar bisa di pakai lagi
@@NIHILProject mayan banyak bang, melalui goggle drive. Udah aku relog dan kosongkan file upload tapi disknya masih saja tidak kehapus
@rizalahmadkusumah7238
2 жыл бұрын
bang yang saya gini, awalnya berhasil, tapi yang kedua kalinya gini, mohon bantuannya RuntimeError: CUDA out of memory. Tried to allocate 794.00 MiB (GPU 0; 4.00 GiB total capacity; 2.35 GiB already allocated; 0 bytes free; 2.96 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
@NIHILProject
2 жыл бұрын
di semua foto kah? resolusinya berapa besar?
@aswinardiansyah3966
2 жыл бұрын
aku pun juga kayak gitu bg, masalahnya ternyata ada di resolusi gambar yg terlalu tinggi. jadi emang gambarnya harus bener-bener buriq🤣
@NIHILProject
2 жыл бұрын
Tergantung memory gpu juga dude, karena saya coba enhance foto burik yang resolusinya besar lebih dari 3000an pixel juga berhasil. Tapi memang umumnya foto burik itu karena resolusinya kecil, kalo reso udah sampe 4K mah udah bisa disebut high-res, berarti emang kang potonya aja yang nubi wk.. (ibarat gear udah pake dslr, tapi foto talentnya blur) Makannya ane juga sering bilang, untuk coba pakai foto lain / perkecil atau kompres ulang (bisa via paint) /discreenshot lagi fotonya.
@indrawijayakusuma5472
2 жыл бұрын
Mantaappp
@amruaj8341
2 жыл бұрын
Dude gw di donload note tutorialnya kok pas d buka bentuk filenya html ya..
@NIHILProject
2 жыл бұрын
Maksudnya XML? file itu bisa dibuka pake notepad / notepad++ juga kok dude
@amruaj8341
2 жыл бұрын
Iya xml maksut slah ngetik trims dude fast responya 🙏🙏 sukses slalu
@amruaj8341
2 жыл бұрын
Dude punyak gw pas instal.pytorch kok eror ya trus hurufnya jd warna merah smua tolong arahanya dude.. 🙏🙏
@davidalexander9935
Жыл бұрын
catatan tutorialnya download dari mana ya dude?
@bimasyachputra4315
11 ай бұрын
bang tutorin dong ubah convert ke exe application gitu
@yosepirwanto7209
2 жыл бұрын
kak ,dibatasin ga, ke 3x kok jd gak bisa error ga ad hasilnya
@NIHILProject
2 жыл бұрын
Gak dibatasin kok, mungkin format fotonya bermasalah / terlalu besar. Coba cek pesan errornya apa, pada step upload/inference.
@bungassafaril
2 жыл бұрын
Bro,gua gagal comnd clan repo. Katanya : git is not recognized as an internal or external comand
@NIHILProject
2 жыл бұрын
Cek deskripsi ya, ada catatan disitu
@bungassafaril
2 жыл бұрын
Okok sip sip 😅😅😅, thanks
@VasiliySlyusar
Жыл бұрын
Thanks for great tutorial. Is it still working ? Because in my case, i was able to process images only on CPU ( maybe there are some changes in software versions ?
@NIHILProject
Жыл бұрын
I tried it again, and it still works on my device. Actually, I rarely use it now because I'm using Stable Diffusion, GFPGAN/Real-ESRGAN, Codeformer, and many more available in the Extra tab of Stable Diffusion. The options are extensive, so I'm using this for upscaling my photos/images now.
@VasiliySlyusar
Жыл бұрын
@@NIHILProject Are you using SD for Windows ?
@NIHILProject
Жыл бұрын
@@VasiliySlyusar Yes, and it is very stable with NVIDIA GPU
@runtorunprojectart
2 жыл бұрын
Bang pakai win7 no VGA Prosesor Radion 7 bisa gk bang 🙏
@NIHILProject
2 жыл бұрын
ane belom nyobain dah kalo untuk win 7, recommended nya win 10 1909 or latest & win 11 dude, dan python 3.9+ keatas juga udah gak support win 7
@Deathshot_Official
Жыл бұрын
Is it possible for ya to make a simple exe/app for windows for GFPGAN?
@NIHILProject
Жыл бұрын
Although I haven't tried it myself, I've seen the community utilize Real-ESRGAN/GFPGAN in third-party photo viewer programs
@Deathshot_Official
Жыл бұрын
@@NIHILProject like? 🥺
@wesleyk.8376
Жыл бұрын
For crap sakes, can't they just put this all into an executable? Damn
@fathurrochman8085
2 жыл бұрын
yang di tunggu2 tqtq bank :v
@NIHILProject
2 жыл бұрын
Moga berhasil dah ya, ini optional aja buat yang ogah buka browser, tapi kalo mau yang gampang tetep via google colab aja ya 🤣
@user-oy4hi2vz5r
Жыл бұрын
bang. spek pc lu apa pas buat video ini
@NIHILProject
Жыл бұрын
Waktu bikin video ini Ryzen 5, Ram 16gb, gpu Gtx 1050 ti
@5minreff
2 жыл бұрын
nyawer ahhhh
@NIHILProject
2 жыл бұрын
uhuuuy👏
@supriyantosupriyanto3610
2 жыл бұрын
Bang klo running tanpa anaconda gimana ya ?
@NIHILProject
2 жыл бұрын
harus tetap mengandalkan anaconda sepertinya, sebenarnya bisa dibuat launcher dengan menggabungkan script ke dalam software photo viewer pihak ketiga, tetapi langkahnya jauh lebih rumit & bisa memakan lebih banyak memori.
@aryanjohnsharma
Жыл бұрын
can this work for people with AMD GPus????????
@NIHILProject
Жыл бұрын
I've read articles about this for suggestions to operate flawlessly with NVIDIA GPUs, however I haven't tried on devices with AMD GPUs. Furthermore, AMD GPU Torch only functions on Linux if Windows can only install CPU. bit.ly/3r7uKC9 Personally, I rarely use GFPGAN because Real-ESRGAN AI can now do the same thing with an additional script "-face enhance" (which is also based on gfpgan) Or else, you can now use the AI on this page (demo online method): replicate.com/xinntao/gfpgan or replicate.com/xinntao/realesrgan. I also discussed it in this video: kzitem.info/news/bejne/2m2HnmhqcaeBd4Y
@aryanjohnsharma
Жыл бұрын
@@NIHILProject what's the easiest way which is totally free ?
@brokenpropfpv8917
2 жыл бұрын
what if i want only the face restoration?
@NIHILProject
Жыл бұрын
Run this script without "--bg_upsampler realesrgan" - before: python inference_gfpgan.py -i inputs/upload -o results -v 1.3 -s 4 --bg_upsampler realesrgan - after: python inference_gfpgan.py -i inputs/upload -o results -v 1.3 -s 4 desc: -i = Input image or folder -o = Output folder -v = GFPGAN model version -s = Upsampling scale of the image -bg_upsampler = Background upsampler. Default: Real-ESRGAN
@brokenpropfpv8917
Жыл бұрын
@@NIHILProject it still upscaled the picture
@Mr.Amatiran
2 жыл бұрын
bang anaconda saya bukan ke buka di system 32 tpi di user
@NIHILProject
2 жыл бұрын
setelah buka anaconda, ubah basenya dengan menulis "cd" seperti contoh di video, yang penting arahkan ke file penyimpanan GFPGAN nya
Пікірлер: 246