DeepFaceLab is the leading software for creating deepfakes.

Create pod from "RunPod Desktop" template (use this link, because they removed the previous option), at least 30GB for volume and container disk sizes, as both will be used to create the environment later.

Click "Connect", "Start Web Terminal" and open a terminal by clicking "Connect to Web Terminal", copy-paste the whole line below, hit enter:

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_23.3.1-0-Linux-x86_64.sh; cd /; sudo chmod 777 /workspace; cd /workspace; git clone https://github.com/DaviSoEditando/DeepFaceLab-Runpod DeepFaceLab_Linux; bash /home/kasm-user/*.sh

Hit the spacebar a few times, answer yes to each question, set install location to /home/kasm-user/miniconda3 (will probably already be), answer yes to the last question and finish the installation.

Quit terminal, open new one - should start with ‘base’, copy-paste the whole line below to create the conda environment that will be used, be patient as it will take a few minutes, hit enter:

cd /home/kasm-user/miniconda3; conda env create -f /workspace/DeepFaceLab_Linux/scripts/environment.yaml; cd /workspace/DeepFaceLab_Linux/scripts/; cp -r /home/kasm-user/miniconda3 /workspace/

Now you can download your model and your videos, just go back to the Runpod page, click on "HTTP Service [Port 6901] Not Ready" which will open a new tab where you will put the login and password:

login:    kasm_user 
password: password

A desktop will open remote and you can download using google chrome and then unzip the zip you downloaded using the command:

unzip /home/kasm-user/Desktop/Downloads/*.zip -d /workspace/DeepFaceLab_Linux/workspace/

cd /workspace/DeepFaceLab_Linux/scripts/; bash 2_extract_image_from_data_src.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 3_extract_image_from_data_dst.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 4_data_src_extract_faces_S3FD.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 5_data_dst_extract_faces_S3FD.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 6_train_SAEHD.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 7_merge_SAEHD.sh

cd /workspace/DeepFaceLab_Linux/scripts/; bash 8_merged_to_mp4.sh

After restarting the pod, just use the command below to create a link from the "miniconda3" folder on the system to get back to using DeepFaceLab again:

ln -s /workspace/miniconda3 /home/kasm-user

But only if you still have the Volume with the previous files.

If you have deleted the volume, you will have to do the whole process from the beginning.

Runpod allows you to create a "Network Volume", in which you can keep the files, both from DeepFaceLab and the miniconda/env installation, that way you are not hostage to the same graphics card, as they tend to get busy quickly, making it very difficult rent again. You will be charged a few cents a month per GB that you have in the generated volume and so far you can only use it in "Secure Cloud" mode.

I got most of these tips from this forum, and spent the night trying to find the causes of the problems that occurred when trying to run the scripts in Runpod.

If you found this tutorial useful, I wouldn't be sad for you to transfer some eth to my wallet.😊

0x0d1F8774a408a3b06DB9aFB75961e5C01D37DE74

Hope this helps!

DeepFaceLab https://arxiv.org/abs/2005.05535 the leading software for creating deepfakes
More than 95% of deepfake videos are created with DeepFaceLab. DeepFaceLab is used by such popular youtube channels as deeptomcruise 1facerussia arnoldschwarzneggar mariahcareyathome? diepnep mr__heisenberg deepcaprio VFXChris Ume Sham00k Collider videos iFake NextFace Futuring Machine RepresentUS Corridor Crew DeepFaker DeepFakes in movie DeepFakeCreator Jarkan
What can I do using DeepFaceLab?
Replace the face
De-age the face
https://www.youtube.com/watch?v=Ddx5B-84ebo
Replace the head
https://www.youtube.com/watch?v=xr5FHd0AdlQ
https://www.youtube.com/watch?v=RTjgkhMugVw
https://www.youtube.com/watch?v=R9f7WD0gKPo
Manipulate politicians lips (voice replacement is not included!) (also requires a skill in video editors such as Adobe After Effects or Davinci Resolve) https://www.youtube.com/watch?v=IvY-Abd2FfM https://www.youtube.com/watch?v=ERQlaJ_czHU
Deepfake native resolution progress
Unfortunately, there is no "make everything ok" button in DeepFaceLab. You should spend time studying the workflow and growing your skills. A skill in programs such as AfterEffects or Davinci Resolve is also desirable.
Mini tutorial
Releases
Windows (magnet link)	Last release. Use torrent client to download.
Windows (Mega.nz)	Contains new and prev releases.
Windows (yandex.ru)	Contains new and prev releases.
Google Colab (github)	by @chervonij . You can train fakes for free using Google Colab.
Linux (github)	by @nagadit
CentOS Linux (github)	May be outdated. By @elemantalcode

Meme zone
#deepfacelab #deepfakes #faceswap #face-swap #deep-learning #deeplearning #deep-neural-networks #deepface #deep-face-swap #fakeapp #fake-app #neural-networks #neural-nets #tensorflow #cuda #nvidia