<aside> 👉 If all of this seems too overwhelming for you, you can simply request models:
How to request a LORA on Discord? Tips and common sense
</aside>
If you have no idea how to train a model, you can focus on creating datasets or tags. This guide will explain in a simple way phases 1 and 2 of the LORA training protocol in Spanish*.*
If you already know how to create datasets and you are just reading this to know how to get the corresponding role to see the channels, it's easy, you just have to go to the #solicita-tu-rol chat and react.
You can use Mega or Google Drive to share your datasets, however, we recommend using the Telegram repository created for this purpose, which is pinned in the "Guide on how to create and send your datasets". Why? For the simple reason that if this server were to disappear, they would still be available on Telegram for all those who join these groups, thus offering us some robustness and resistance to censorship.
<aside> 🆕 If you want your concept to be trained as Lycoris, notify the person who will start training it in the thread you have created, otherwise, it will be left to their discretion.
</aside>
Add "No tags" if your dataset has no tags, and "free" if no one has requested it for training yet. If it has tags, as it is obvious, add the tag "With tags". If it is in the training process, add "In training".
What is dataset creation?