AI for Architects:
LoRA Fine-tunes
Are you an architect or designer eager to delve into the world of AI and fine-tune models to fit your specific needs? Look no further! Our workshop, "AI for Architects - LoRA Fine-tunes," is tailored just for you.
In this workshop, we will explore the fundamentals of LoRA fine-tuning, a method that allows for precise adjustments to pre-trained AI models to better suit specific design requirements. LoRA provide a way to enhance the capabilities of Stable Diffusion models, making them more adaptable to unique architectural style. We'll cover the basics of LoRA theory, dataset preparation, and the use of KohyaSS UI for training. Participants will learn how to navigate training settings, define training parameters, and refine models using Automatic1111.
Through hands-on exercises, you'll gain practical experience in fine-tuning models, testing outcomes, and applying these techniques to your design workflow. By the end of the workshop, you'll have a solid foundation in LoRA and a clear understanding of how to leverage these techniques to enhance your AI-driven designs.
Join us to master LoRA fine-tuning and take your design process to the next level.
Workshop Overview
AI for Architects - LoRA Fine-tunes is an in-depth workshop tailored for architects and designers looking to master fine-tuning techniques for AI models. This session will cover the fundamentals of LoRA fine-tuning, dataset preparation, and training settings, providing participants with the skills to customize AI models for their specific design needs.
Why it is important
Fine-tuning LoRA models is essential for creating custom solutions that meet specific design requirements. Understanding how to use techniques like LoRA fine-tuning is crucial for staying competitive and leveraging the full potential of AI in architecture. This workshop provides the foundational knowledge necessary to integrate these advanced tools into your design workflow, making the transition to AI-driven design seamless and effective.
What will you learn
In this workshop, you will learn the theory behind different fine-tuning techniques and their main differences. You will gain hands-on experience with dataset preparation, including image selection and captioning, and learn to navigate KohyaSS UI for setting training parameters. The workshop will cover practical exercises in training settings, testing, refining, and LoRA mixing. By the end, you'll have the skills to seamlessly integrate finetuned AI models into your design workflow, enhancing creativity and efficiency.
Mentors
Ismail Seleit
Ismail Seleit is a design architect at the industry-leading Foster and Partners, specialising in design technology. With extensive experience in architecture and computational design, Ismail has contributed to diverse projects, ranging from design competitions to the realisation of buildings on various scales. Actively supporting design teams, Ismail leverages his expertise in BIM and Computational Design to navigate complex challenges.
His primary focus is on enabling project teams to achieve efficient, informed, and collaborative design outcomes. Collaborating across departments, Ismail works to implement innovative methodologies to manage the design process from urban planning to product scale.
Simultaneously, Ismail engages in applied research, exploring new opportunities for implementing cutting-edge design workflows. His research specifically delves into the integration of generative AI image-generating tools, with a keen interest in the combination of Stable Diffusion and ControlNet.
This unique approach offers various techniques for generating controlled image compositions based on diverse user-defined inputs. Beyond his architectural pursuits, Ismail is an ambient-electronic music producer, with a focus on film scores.
He has composed and produced soundtracks for architectural short films at Foster and Partners and contributed to the music and sound design of various independent films. Ismail’s passion for creative thinking, coupled with his dedication to design problem-solving, continues to drive his multifaceted and innovative endeavours.
Programme
Introduction
- The Ismail Seleit secrets and his journey with LoRA Fine-tunes
- LoRA theory, different methods and main differences
- SD 1.5 vs SDXL
Dataset Preparation and Captioning
- Dataset preparation: image selection
- Dataset preparation: captioning
Training with KohyaSS UI
- Intro to KohyaSS UI
- Training settings
- Training parameters definitions
- Setting training parameters using my preferred workflow
Refining and Applying AI in Design
- Testing and refining in Automatic1111
- LoRA mixing
- Best practices, closing thoughts
Important info
Participants will need access to Stable Diffusion, either installed locally or using an online instance like RunDiffusion: https://rundiffusion.com/. The workshop will be recorded, and all participants will receive access to the recordings.
Important Info:
- It is advisable to have active plans for the following tools: Midjourney, Prome AI, Krea AI and Runway. This tools covered have free versions, except for Midjourney.
- Basic knowledge of interior design and architecture is recommended.