From the course: Complete Guide to Data Lakes and Lakehouses

Unlock the full course today

Join today to access over 25,200 courses taught by industry experts.

Project execution: Using the copilot

Project execution: Using the copilot

From the course: Complete Guide to Data Lakes and Lakehouses

Start my 1-month free trial Buy for my team

Project execution: Using the copilot

“

- [Instructor] In order to execute our generative AI application and actually get it to run without melting our computer or codespace, we need a GPU. At the moment of recording, GPUs are not available in codespaces, but we can use a Google Colab notebook, which offers a few GPU compute units for free, so we can run our Ollama models there. All you need is a Google account. Just so you know, gen AI applications are heavy. So the free compute units are limited. I was able to get a few questions before I ran out of credits and I needed to get more. It's not really expensive, but it's good to know. So I'm going to get out of full screen now and go to colab.research.google.com in my browser and create a new notebook there. Click on create new notebook. I added the contents of the notebook to the GitHub repo so you can copy it from there. It is under the notebooks folder in a file called run_ollama_gpu. So I'm going to copy the contents from here. And it's only two blocks, so it's really…

Contents