Universal Ai Diploma Week 5 Generative Adversarial Neural Network Instructions

God Bennett
3 min readFeb 11, 2022

1. In chrome, add metamask extension, and create address.

2. Create nft account on async art. (Nfts are like digital art unique by time and author. You can copy mona lisa art, but it will always have an original copy/author. An original NFT though digital, can be unique in time by an author.)

3. Use the gan colab instance, to create a custom art piece.

a. Go to google and grab 2 images, the *target* image you want to re-style, and the *source* image you want to use as your styling source. (In google find any *target* image and any *source* image and copy image address with file name extension!!)

Ensure link has name extension file like “.jpg” or “.png” etc at the end!

b. Go to section “Download images and choose a style image and a content image:” and make the changes below:

4. Run cells up to section “Now run a few steps to test:’ to see your custom image.

5. Upload your image to Async Art, by following stepes below:

Next Custom/Custom, then Setup Layers by adding images (just add the same image many times, but ensure to copy the image and have the image twice with different file names!!!!!).

Also ensure all data/fields are filled including description.

Add another layer with the same image, following the constraints above.

Add cover images, by taking your custom image, and ensuring it has the same with and length, then saving that as yourimageCover.png.

Next give your project a title and description

Click next, then click test your piece to ultimately obtain the url which everybody else can view.

Get your link and create a repository with the link below as seen in the sample:

https://async-explorer.herokuapp.com/test/canvasID=62067f4a7b2d8a4ca9bd2aed

G. Bennett (Kingston), Artificial Intelligence Lecturer of Universal Ai Diploma at ASTIT & Software Developer | Instagram/G_Bennett | Github/G_Bennett

--

--

God Bennett

Lecturer of Artificial Intelligence, and inventor of “Supersymmetric Deep Learning” → Github/Supersymmetric-artificial-neural-network