100 Days of Gen AI: Day 30

I read lots about fine-tuning today. I started with the OpenAI Fine-Tuning documentation, then branched out into several blog posts and perused many datasets on Hugging Face. My goal is to try fine-tuning as a learning opportunity. I’d like to work with a dataset small enough that I can run the fine-tuning training on my (brand new!) M4 laptop.

Some fine-tuning ideas could be train a model to:

• generate text in my personal voice and style
• recognize my face in a photo
• recognize a product from a photo
• what are the chances I will like a given article of clothing based on its characteristics (fabric, price, country of origin)

Just some ideas, none of them original, but perhaps useful as a learning exercise.