From the course: What Is Generative AI?

Creating your own content - DALL-E Tutorial

From the course: What Is Generative AI?

Creating your own content

“

- So now that we have our generative AI model and our chassis, we are ready to create our own content. If we're a beginner, we can use a paid service like Midjourney or Lensa. If we are more experienced, we can use a notebook and pick from available models. If you download a commercial app and upload only 10 pictures of yourself, like I did, the app suggests a variety of different avatars of yourself. It's super fun. If you are a more experienced generative AI user, a creative technologist, you can go to GitHub, choose your favorite generative AI model, and see if it is available in the form of a notebook. If it's not available, you can still inquire about it inside the generative AI community. They're super friendly people. And if you are a programmer, you can also create your own notebook by taking the model code from GitHub. For a demonstration, we will be running a Google Colab notebook named Deforum, that is based on stable diffusion, to generate a fantasy landscape. As you can see here, the notebook runs through the code, and depending on your settings, produces a personalized outcome. Google Colab requires a subscription. If you would like your generation time to be minimized you need to buy a paid subscription. These notebooks offer a lot of personalized options for the quality of your outcomes. You can always choose the default generation, but the beauty of working with a notebook is that you can tweak and personalize the outcomes. In summary, a model is a set of algorithms that have been trained on a specific dataset. A notebook is a tool for writing and running the code. A creative application is an example of how a model can be used, and the generated outcome is what the end user produces by using a generative AI service or a notebook that houses the model inside it.

Contents