<h1>Lazy-ass low-effort LoRA (LALELoRA?) </h1>DISCLAIMER: I hate this way of making LoRAs. It is effective, it is fast, it is a mess. I can spot a few LoRAs on Civitai which were done this way.This is how I made <a rel="ugc" href="https://civarchive.com/models/59574/kcutetanks">kCuteTanks</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/60284/kcutecreatures">kCuteCreatures</a>, <a rel="ugc" href="https://civarchive.com/models/59755/khorrorcrabs">kHorrorCrabs</a>, <a rel="ugc" href="https://civarchive.com/models/60553/kvoidenergy">kVoidEnergy</a>, and <a rel="ugc" href="https://civarchive.com/models/60640/khastur">kHastur.</a> It takes 5 minutes going from idea to clicking the "train" button.REQUIREMENTS: Midjourney, Python, 5 minutes of time Minute ONE (data selection)<ul><li>Open Midjourney, select explore, write your idea (like "cute creature"), hit Enter.</li><li>Scroll down, click "expand results".</li><li>Click "enable selection" in the top right corner, then click "select all" in the bottom.</li><li>Optional: repeat with variations (like "cute animal")</li></ul>Minute TWO (data collection)Click "Downloader" and download all. Unzip all files to a single folder.Wait a minute -- we didn't even check the pictures. What if there is a picture which does not fit the idea?<ul><li>First, we can't wait a minute. We have to make it under five.</li><li>Second, do not care, this is the lazy-ass low-effort approach.</li></ul>Minute THREE (captioning)The names of the downloaded files have the following form: [user]_[partof_prompt]_[id].pngGuess what: we just take the middle part and make it a caption.We can also add a trigger word, and remove some words, like "cute", "creature".GPT can easily write a procedure, which will do it for you. Since Civitai needs a file uploaded, I'm putting a python procedure I used.Wait a minute -- The procedure does not ensure removal of the complete username. Also, there is only a part of the prompt. Such captions will be a mess!<ul><li>First, we can't wait a minute. We have to make it under five.</li><li>Second, do not care, this is the lazy-ass low-effort approach.</li></ul>Minute FOUR (training parameters)Just take the config for kohya-ss from the other tutorial (<a target="_blank" rel="ugc" href="https://civarchive.com/models/52697/tutorial-konyconi-style-lora">https://civarchive.com/models/52697/tutorial-konyconi-style-lora</a>), set the number of epochs to 1-4 (depending on the amount of pictures), and click the "train" button.Wait a minute -- Shouldn't we fiddle with the parameters, find what is best for the specific LoRA we are making?<ul><li>First, we can't wait a minute. We have to make it under five.</li><li>Second, do not care, this is the lazy-ass low-effort approach.</li></ul>Minute FIVEOh, there is no minute five. We actually could wait a minute.

Files with SHA256: 865451dd4663a173a76b04c19e06e878f28b112868438db9ed285b764d2eccb8