Author: nanonomad

Profile picture of nanonomad
Website
http://nanonomad.com

Posts by nanonomad

Training SDXL to Generate Text Using IA3 LoRA

Posted on

Resources: Test IA3 Text-Image Adapters:https://huggingface.co/AOLCDROM/iA3-adapters More about the Prodigy optimizer:https://github.com/konstmish/prodigy More about the IA3 PEFT adapter:https://huggingface.co/docs/peft/conceptual_guides/ia3 T-shirts dataset:https://www.kaggle.com/datasets/sunnykusawa/tshirts More on Pangrams:https://www.prdaily.com/16-clever-pangrams-for-word-lovers/ Font File to .PNG Image Script: […]

Read More… from Training SDXL to Generate Text Using IA3 LoRA

Replacing Faulty Asus Phoenix RTX 3060 GPU Cooler – It’s Easy

Posted on

This video is about replacing a faulty fan on an Asus Phoenix RTX 3060 GPU. I was experiencing an increasingly loud cicada-like shrieking noise coming from the fan. I decided to replace the fan instead of trying to fix the damaged one with compressed air, because compressed air rarely works well enough. The video goes […]

Read More… from Replacing Faulty Asus Phoenix RTX 3060 GPU Cooler – It’s Easy

Bark Voice Cloning, TTS, RVC, Music Generation and More with the TTS Generation WebUI

Posted on

A look at the TTS Generation Web UI for Bark text to speech, generating music, translation and more. Just a broad overview of what seems to work well, and not so well in this feature-packed project. Video Link: https://www.youtube.com/watch?v=Y8J717tr9t0 Sources for RVC Models: https://rvc-models.com/ https://voice-models.com/ https://huggingface.co/spaces/zomehwh… TTS Generation WebUI: https://github.com/rsxdalv/tts-generation-webui […]

Read More… from Bark Voice Cloning, TTS, RVC, Music Generation and More with the TTS Generation WebUI

Train Better Stable Diffusion Models and Prep Datasets Using this Free “Magic” Image Tool

Posted on

A few tips on using the Paint 3D tool, included with Windows 10, to touch up/edit images for image training datasets. I go over some of the things that I would focus on removing before training a Stable Diffusion LoRA. The Magic Select tool makes it a breeze, once you get a hang of what […]

Read More… from Train Better Stable Diffusion Models and Prep Datasets Using this Free “Magic” Image Tool

Emulate a Sound Blaster in real MS-DOS on Modern Hardware | Retro Gaming on “Current” PCs

Posted on

A look at a little lightweight Lenovo Thinkpad X61 from around 2007, and how it performs running a few games on MS-DOS with sound provided through SBEmu, a Sound Blaster emulator. This laptop has a fickle AD1984 audio chipset (carrying the Intel HD Audio label), and doesn’t have good OS support anywhere; but it does […]

Read More… from Emulate a Sound Blaster in real MS-DOS on Modern Hardware | Retro Gaming on “Current” PCs

How to Play Hundreds of Point-and-Click Adventures on iOS for FREE with ScummVM with NO SIDELOADING

Posted on

Play hundreds of classic PC games like Kings Quest, Quest for Glory, Day of the Tentacle, Full Throttle, and oldschool text adventures on iOS/iPad OS for free with ScummVM. I’ve been playing a lot of classic adventure games on ScummVM lately. If you haven’t heard about it yet, ScummVM is now on the Apple App […]

Read More… from How to Play Hundreds of Point-and-Click Adventures on iOS for FREE with ScummVM with NO SIDELOADING

Training LoRAs and GLoRAs for Stable Diffusion 1.5 and XL Using the New Prodigy Optimizer

Posted on

Training LoRA and GLoRA on SD 1.5 & XL with the Prodigy Optimizer using the Kohya_SS scripts.In today’s video I look at training LoRA and GLoRA adapters for Stable Diffusion 1.5 and XL using the Prodigy optimizer on a large and varied dataset made up of 16 characters. Then I show an example of how […]

Read More… from Training LoRAs and GLoRAs for Stable Diffusion 1.5 and XL Using the New Prodigy Optimizer

Site will be updated/worked on soon

Posted on

I’ve been dealing with some health issues, so the site remains unfinished. Videos are on pause for now, but I’m still working on a few things.In addition, my PC finally died from the load of training ML models nearly 24/7 for the past year. Either the CPU or MB is completely dead. No visual signs […]

Read More… from Site will be updated/worked on soon

Fine Tuning Mistral 7B

Posted on

Can you train new or forbidden knowledge into a LLM? Let’s fine out as I throw 1 gigabyte of scraped, cleaned, plaintext KiwiFarms posts at Mistral 7B. I go over my experience fine-tuning Mistral 7B on a few large datasets of scraped text data including English language song lyrics, and a huge KiwiFarms post dataset. […]

Read More… from Fine Tuning Mistral 7B