Skip to content

e2ckp/DeepFaceLab_Colab-1

 
 

Repository files navigation

Good News

Colab upgraded from the original K80 12G to T4 16G, faster, more VRAM!

Overview

The purpose of this project is to provide a way to run DeepFaceLab for free.
When you have learned how to use DFL, Perhaps the biggest limitation for you is the Computer performance.
Highly equipped graphics cards are very expensive, and cloud services are not cheap.

So how free? take this!^_^ Of course there are some restrictions

The code comes from DeepFaceLab and DeepFaceLab_Linux .
This rep just add the .ipynb file ,makes it simpler!
No need to copy, zip, unzip multiple times, no need to worry about data loss.

Features

  • You can use tesla T4 for free

  • Your data is saved in the google Drive

  • You can preview the results online,No need to download

  • Colaboratory requires interaction when in use. The system may stop long-running background calculations

  • Sometimes,The first load(train load) may be slower .

Guide

  1. First, You need a google account
  2. Secend,you need a DeepFaceLab_Colab.ipynb file
  3. Third, Click Open in Colab ,Run it!
  4. Fourth, Run ViewLastHistory.ipynb file,Check out the latest results.
  5. Sixth,Continue training Step 6 of using the DeepFaceLab_Colab.ipynb file!

About TimeOut

https://research.google.com/colaboratory/faq.html#drive-timeout

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Jupyter Notebook 71.9%
  • Python 27.3%
  • Shell 0.8%