Hi everyone, I’m completely new to object detection and coding in general, but I’m very interested in learning more about it. I’d like to try out a pre-trained RR-DETR (nano) model on a Google Colab notebook, I trained it on roboflow, but I have no idea how to upload it and get it running on google colab
Could someone please help me with a simple guide or example to load and test it in Colab? Even a very basic step-by-step with code would be super helpful. Thanks in advance!
You’re going to have fun with all this stuff! If you’re new to every aspect of this, I’d say you’d have more success building a workflow where you can then test your model right in the browser. Roboflow seemed to have created the Workflow feature for just this situation! This article will help you get started. If there’s a reason you want to work in Colab maybe expand on why that is critical and others can help with that aspect. Happy building!
Hi @giorgio!
First off, welcome to the Roboflow Community!
This is a great question! To test your model in Colab, start out building a workflow similar to the one below:
Next, click on the Deploy button within the red circle above. This will give you the code snippet seen below, that you can copy and paste into Colab to test your model!
Hi @Ford,
Thanks so much for your reply — images are working great now, really appreciate it! I also tried using videos, but I seem to be running into a wall there with google colab. I don’t know how to do it with videos. I tried running it on my PC as a server, downloaded Docker and Roboflow Inference, but it seems like neither of them is working — due to a weak setup. What should I do now?
Thanks a lot! I’ll definitely check out the Workflow feature. I was using Colab mainly because my local setup isn’t strong enough, especially for working with videos. But if the browser-based option supports that too, it might be the better way to go. Appreciate the help!
