Skip to content

subhyde/DeepLearning-AI-Images

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 

Repository files navigation

About this project

This was a project created to experiment with machine learning and GAN image processing.

Unfortunately I don't have an nvidia gpu and AMDs support for machine learning is next to none.

What I did was train the model with images of myself and with the help of a discriminator running on my cpu. I had to utilize checkpoints since training a neural network on a CPU alone does take some time

Here are some of the best results that I have gotten from my neural net

generated_img_000_2+new111 generated_img_003_1+new111 generated_img_006_4+new111

Once tensorflow and AMD start supporting their 6800xt lineup i will definately be revisiting this project and experimenting with other types of machine learning branches and push the limits of what can be done

If you would like to try this project for yourself you will need to install

  • TENSORFLOW

  • KERAS

and create 3 directories in the root of the project

  • training_dataset/images -- you will put all of your images in this directory

  • training_model/checkpoint -- this will be your checkpoint directory

  • created/images/ -- this will be where all of the gererated images end up after each epoch cycle

if you are struggling to the the program to run, take a look at the official documentation here https://www.tensorflow.org/tutorials/generative/dcgan

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages