Walkthrough on Fetch.ai’s CoLearn pAInt
Aug 24, 2021
A brief explainer on how the NFT platform will work and how you can participate in it
Welcome to Colearn pAInt. This guide will help you understand how the platform works and how you can create art. We will keep it very simple and easy to understand.
Step 1 = Connect your metamask with the platform.
Keep sufficient FET and ETH for gas fees.
Step 2 = The BID PHASE
Be ready to place a bid. Word of advice — if at any point your metamask is slow to approve the fees, you should check the network you are connected to and that you’ve enough FET and ETH for the gas fees.
At a Dutch Auction, prices start high and are dropped successively until all the slots are sold. Once all the slots are sold, the auction ends.
The aim of the dutch auction is to allocate all the slots for participating in the fetch.ai collective art event.
The auction starts at a specified block height and runs for a defined number of blocks. The asking price starts high and then decreases linearly until the end of the auction. So if the asking price starts at 9000 $FET and ends at 1000 $FET then half way through the auction the current asking price will be 5000 $FET.
Auction round lasts 24 hours — within the 24 hours, you can bid at any point, but your bid needs to be over the current asking price. So if the current asking price is 5000, then 5000, 5001, 7000, 10000 and 13243 would all be valid bids. But anything less than 5000 would be an invalid bid and would be rejected.
Once the auction closes, the price of a slot is finalised. Then the number of slots that each user gets is finalised. The bidders are allocated slots based on the amount of their bid divided by the final slot price.
No. of Slots = [Amount of your bid]/[Final slot price]
There’s only a fixed number of slots, so the slots might run out before they get the later bidders. To see how this works, look at the following example.
Imagine that the final price of a slot is 1000, and there are ten slots, and the following bids were entered:
b1: 4000 (The first bidder would be allocated 4000/1000 = 4 slots)
b2: 5500 (The second bidder would get 5500/1000 = 5 slots and 500 $FET leftover)
b3: 4000 (But now there’s only 1 slot remaining, bidder 3 gets it, with 3000 FET leftover)
b4: 1000 (the fourth bidder wouldn’t get any slots)
STEP 3 = The BUILD PHASE
The build phase is live for 24 hours. It is quite easy! You remember MS Paint? Yes, you just use that white pad to scribble some lovely art work of your choice until it reaches 100%.
It can be something as simple as squiggly lines or you could also create a masterpiece as shown below. Once you’re done, click on Submit Entropy. If you get an error message “Failed to submit entropy”, just refresh the page and try again.
STEP 4 = The VOTING PHASE
- There are 3 voting rounds. Each voting round will last for 24 HOURS.
- In each round, you must choose 5 images which you can either drag and drop or you can simply check the white box on the top right.
- Submit the vote once you’re done and wait before the next round begins.
The first voting round is called Shaping the Artwork — this is where you start describing the patterns for the machine learning algorithm to learn and analyse from.
The second voting round is called Refine the Artwork — this is where you further refine the process for the ML algorithm to create the final art pieces.
The third voting round is where you select your preferences from the images generated by the ML algorithm.
Step 5 = The Winners
That’s it. Here are the winners of the voting round and when you click on the names underneath the art, you will be directed to the OpenSEA platform where once the art is sold, you can receive the proceeds from it depending on the slots you have.
Once this is complete — do not forget to collect your leftover $FET from the bidding process on the top right corner. When the transaction is successful, you will see a message on the bottom left corner “Successfully withdrawn”
We hope you enjoy the platform! This is a first of its kind art generation platform using the collective learning framework from us at Fetch.ai and we are very proud to share it with the entire crypto community.